Physical AI Brief
Daily cross-source signals for the Physical AI supply chain — silicon photonics, CPO, VLA models, humanoid hardware, embodied AI. Three streams, one page, zero filler.
381 items today · 317 arxiv · 2 SEC 8-K · 62 humanoid · 0 CN photonics
01 ARXIV · PHYSICAL AI PAPERS
317 items- arxiv:2610.31670 · cs.ROSensorimotor Alignment: A Platonic Proxy for Embodied Policy SuccessYunfei Xie, Howard H. Qian, Shiyi Lan, Kaiyu Hang +1
Robotic foundation models are typically evaluated through costly closed-loop rollouts, which motivates cheap offline proxies for model capability. The Platonic Representation Hypothesis (PRH) suggests that as representations improve, different modalities such as vision and language converge toward a shared statistical model of reality. We ask whether motion trajectories form another such modality: as embodied models become more capable, should their sensory representations converge toward the structure of physical motion? To examine this, we propose Sensorimotor Alignment ($S^2$), a training-free offline metric that measures how well a model's sensory representation aligns with the geometry of physical motion. We find that $S^2$ correlates positively with downstream robot task success across embodied model families, making it a low-cost offline proxy for prioritizing models before closed-loop evaluation. $S^2$ can also identify promising vision-language model (VLM) backbones even before costly fine-tuning into vision-language-action models (VLAs). Beyond this correlation, $S^2$ offers representation-level explanations for phenomena observed in VLAs and world action models (WAMs). With $S^2$, we extend PRH to embodied models and make sensorimotor alignment empirically measurable, offering insight into why an embodied model succeeds beyond its closed-loop success rate.
vision-language-actionembodied - arxiv:2610.31668 · cs.CVReading, Writing, Sharing, and Interacting with the World through VideoShengqiong Wu, Meng Luo, Bobo Li, Ziwei Liu +4
Video foundation models have advanced rapidly in understanding and generating visual sequences, while video world models seek to capture the underlying dynamics of the world they depict. Despite their different objectives, both concern the relationship between video observations and an evolving world. This survey adopts world modeling as a unifying lens for examining these developments. Viewing an observed video as a partial, lossy, and temporally ordered projection of a latent temporal world, we organize the literature around four complementary operations: reading infers world properties from observed video; writing synthesizes video from specified or learned world conditions; sharing integrates understanding and generation through common modeling mechanisms; and interacting predicts and updates world states under actions and evolving observations. This perspective distinguishes temporal coherence in observed or generated sequences from consistency in the underlying world state. Across tasks, representations, models, datasets, and benchmarks, we trace a shift in emphasis from clip-level recognition and perceptual fidelity toward persistent state, controllable evolution, physical and causal validity, and evaluation grounded in interaction and downstream decisions, and we distill the open challenges that follow.
world modelpersistent statebenchmark - arxiv:2610.31662 · cs.LGReliable AI-Assisted Decisions via Selective AuditingIsaac Gibbs, Elise L. Han, Ryan J. Tibshirani, Tijana Zrnic
We consider online monitoring of an AI system that produces a sequence of decisions. At each round, the system outputs a proposed decision together with a confidence score estimating the trustworthiness of the decision. The decision and confidence score need not be produced by the same model: for example, one agent may propose a decision, while a separate verifier agent evaluates it and outputs a correctness or safety score. Based on this information, we must decide whether to accept the AI decision automatically or consult an expert, such as a human auditor, who reviews the proposal and either accepts or corrects it. We formulate this task as an online constrained optimization problem in which the goal is to minimize the frequency of expert auditing while guaranteeing that the fraction of AI errors that go unaudited stays below a user-defined threshold. We derive a family of algorithms with provable error-control guarantees with no assumptions, including a method that is optimally efficient: it asymptotically attains the best achievable auditing rate for both adversarial and i.i.d. data streams. Experiments on screening harmful requests to large language models, qualitative text annotation, and data labeling show that our methods achieve valid error control in practice while substantially reducing the need for human auditing compared to baselines.
agent - arxiv:2610.31661 · cs.ROOpenTac: An Open Ecosystem for Customizable, Scalable, and Transferable 3D Tactile SensingAriel Slepyan, Geeta Chandra Raju Bethala, Jiafeng Gu, Anagha Rajendra Dangle +11
We present OpenTac, a novel sensor and a full-stack ecosystem for tactile skins that are highly customizable, scalable, and readily deployed in downstream robot control and learning applications. Our core contribution is twofold. First, we develop a scalable resistive tactile skin with a structured deformable overlay that converts tangential loading into spatially varying pressure. Differential responses across neighboring "taxels" support decoding of local shear vectors, and the common-mode response encodes normal force. Second, we provide a highly customizable sensor design tool that is able to automatically generate fabrication files for these tactile skins from 2D outlines and developable 3D surfaces, including from abstract descriptions such as text (e.g., "design a sensor for a size 9 shoe insole"). We develop an ecosystem around the OpenTac sensor, including the CAD design tool, a sensor simulator, cross-sensor transfer, and fast/adaptive data acquisition. We demonstrate the broad applicability of the OpenTac-CAD sensor design tool by designing and fabricating more than 30 sensors across a variety of robot embodiments. We characterize the three-dimensional tactile fields produced by OpenTac with a commercial, high-precision force/torque sensor. We demonstrate the broad applicability of 3D tactile sensing across contact-rich feedback control and manipulation policy learning on four real-world robot embodiments. Additionally, adaptive acquisition enables kilohertz-rate partial readouts, exceeding full-array raster scanning on the same sensor by 19.2X at 16 sampled taxels per update for a 1024-taxel skin. We provide all hardware, software, and result videos are at opentac.github.io
manipulationtactile - arxiv:2610.31657 · cs.LGCost-Aware Abstention for LLM-Based IT Ticket ClassificationSachin Gupta
Wrong automatic routing of an IT support ticket can cost more than human triage. We evaluate cost-aware abstention as an offline, cost-conditional readiness assessment: select a routing policy on calibration data under explicit costs, then evaluate that frozen policy once on held-out test data. On a public 47,837-ticket, eight-class corpus, we compare four trained non-LLM classifiers with four large language models under zero- and few-shot prompting. Under the illustrative cost model (c_ok, c_triage, c_wrong) = (0, 1, 5), TF-IDF with logistic regression and a threshold of 0.60 selected on calibration reduces test cost from 0.763 to 0.472, a 38.1% relative reduction (paired ratio-bootstrap 95% CI [35.7%, 40.4%]), while lowering wrong routes from 15.3% to 3.5%. Frozen calibration-selected policies also beat both always-route and always-defer on held-out test data in 14 of 16 settings in a post-hoc cost-parameter analysis. The strongest prompted configuration reaches 0.477 accuracy on a stratified 1,000-ticket test subset, versus 0.879 for fine-tuned DistilBERT on the same tickets. Isotonic lowers expected calibration error without reliably improving discrimination: GPT-4o-mini few-shot has one confidence value before and after recalibration and correctness AUROC 0.500. We therefore report correctness AUROC and tie-aware risk-coverage area alongside ECE. At the analytically derived break-even probability 0.80, seven prompted configurations defer all tickets; Haiku few-shot routes only 19/1,000. An LLM-only agreement control and margin/entropy rejection baselines further separate calibration from discrimination. The findings are scoped to this dataset, prompt design, model endpoints, and illustrative costs; the work is an evaluation framework, not a new abstention algorithm or universal deployment certification.
evaluation framework - arxiv:2610.31656 · cs.LGMETTA: Memory-Enhanced Test-Time Adaptation for Meta-Reinforcement LearningAli Najar, Hosein Hasani, Amir Kooshan Fattah, Yasamin Rajabi +1
Meta-reinforcement learning aims to train agents that adapt quickly to unseen tasks from limited interaction, but this requires representing task evidence over long interaction histories. Classical recurrent agents compress experience into hidden states, while transformer-based agents retrieve from past trajectories at quadratic memory and computation cost. In this work, we propose Memory-Enhanced Test-Time Adaptation (METTA) for meta-reinforcement learning. It decomposes the fast adaptation phase into two temporal scales: a causal neural memory performs fine-grained within-episode adaptation with linear sequence complexity, while compact episode-level summaries support long-range cross-episode task inference without attending over the full interaction history. Moreover, equipping agents with neural memory provides sufficient expressive capacity and test-time training capability, enabling effective adaptation to unseen held-out tasks and task families at meta-test time. On unseen held-out Meta-World ML10/45 test environments, METTA achieves a success rate over $50\%$ while maintaining a low training budget and short wall-clock time. On long-context sparse-reward PointMaze tasks, it outperforms transformer-based baselines by exhibiting more structured exploration behavior. These results show that METTA provides an efficient path toward more adaptive and scalable long-context meta-RL.
memorylong-context - arxiv:2610.31651 · cs.CVBrickWright: Agentic Creation of Physically Feasible Brick BuildsAva Pun, Kevin Dai, Sally Chen, Maxwell Jones +6
Frontier coding agents are becoming increasingly adept at 3D modeling. Yet, they still face challenges when building objects meant to be assembled in the physical world, often creating invalid assemblies with floating, colliding, or disconnected parts. We argue that providing agents with tools for assembling builds and verifying their physical feasibility is sufficient for them to produce physically valid assemblies that still adhere to input prompts. Focusing on the domain of LEGO brick builds, we present BrickWright, a framework for generating physically feasible brick builds from text or image prompts using pre-trained agents, where a build is physically feasible if it uses only real parts, has no colliding or floating parts, and forms a connected structure if possible. BrickWright guides agents to create builds by writing code with BrickLayer, a Python toolkit with two main components: an ergonomic brick assembly API that halves inference cost compared to working in raw LDraw, and a validator that provides feedback on physical feasibility. Whereas baseline agents achieve physical validity rates ranging from only 9% (GPT-6 Luna) to 69% (GPT-6 Astra), BrickWright-guided agents achieve 100% validity on a sample of 512 prompts, while improving prompt adherence for lightweight agents and maintaining it for stronger agents. Additionally, they outperform specialized models on the text-to-brick task. Code is available on our project page: https://avalovelace1.github.io/brickwright/
agentic - arxiv:2610.31650 · cs.LGEpiSEM-Bench: Evaluating Safety and Evidence Recovery in LLM Agents under Missing EvidenceQijin Zhang, Hanzhen Wang, Yuehan Zhao, Haoyu Dai +3
Tool-using language model agents rely on environmental facts to decide whether an action is permitted. When a tool response omits a safety-relevant fact, an agent may carry out a prohibited action even though the user request is benign and every returned statement is true. We introduce \bench{}, an executable benchmark for evaluating unsafe execution, safety boundaries, and evidence recovery in LLM agents under missing evidence. It contains 300 matched unsafe/safe task pairs across six domains. By systematically suppressing critical facts, the benchmark measures unsafe execution under critical evidence omission and determines the minimum observed failure cardinality, including failures with full evidence and cases with no observed failure after exhaustive testing. An independent audit tool tests whether agents seek missing facts and then withhold prohibited actions while completing permissible tasks. The benchmark reveals substantial variation across the ten models in omission robustness, evidence acquisition, and appropriate action. Resistance to missing evidence does not necessarily translate into useful recovery. Recovery access reduces unsafe execution for all ten models, but agents differ in whether they actively obtain the missing facts and use them to act appropriately in both worlds.
agentllm agentbenchmark - arxiv:2610.31649 · cs.CVGate3DGen: A Pre-processing Model that Selectively Purifies Real-World Images for 3D GenerationXiaolin Li, Zihao Huang, Fanpeng Meng, Chen Yang +4
Single-image 3D generation has achieved high fidelity on curated benchmarks, yet degrades sharply on real-world images with uneven illumination, blur, and occlusions. Purifying such inputs requires deciding when to intervene and preserving the visual cues needed for 3D generation, while leaving clean inputs unchanged. We introduce Gate3DGen, a selective pre-processing model that gates image inputs to existing 3D generators without retraining them. A degradation-aware router bypasses inputs classified as clean and extracts features to guide purification of the remaining inputs. Conditioned on these features, a 3D-oriented one-step purifier purifies degraded inputs with spatial losses that encourage preservation of object identity, contours, and textural cues. To provide paired supervision, we construct a synthetic data engine that renders clean views of 3D assets and synthesises composite degradations. On the synthetic benchmark, the purifier improves image fidelity and downstream 3D geometry in only 2.3s on average, reducing TRELLIS.2 mesh distance by 61.5% relative to degraded inputs. VLM evaluation and human preferences on real photographs further support identity preservation and degradation correction. More demos and code are available at https://supra-3d.github.io/gate3dgen.
benchmark - arxiv:2610.31646 · cs.AILow-Bit N-Gram Embeddings for Language ModelsStephen Chung, Wenyu Du
Many recent open-weight language models use n-gram embedding tables to improve performance. These tables store learned representations of short token sequences, such as pairs or triples of consecutive tokens, which the model retrieves during processing. However, they can require substantial memory: DeepSeek-V4.1-Flash, for example, allocates 196 billion parameters to its Engram n-gram memory. We therefore investigate whether this memory cost can be reduced while preserving the performance improvement that the tables provide. We propose storing n-gram embeddings at low precision and systematically investigate different quantization methods and bit allocations across tables. Across nanochat-, Llama-, and GPT-2-family models trained from scratch, four-bit tables save approximately 75% of table memory while preserving more than 99% of the validation-loss improvement from adding the tables. Using fewer bits leads to a larger drop in performance, although we find that a short fine-tuning phase recovers a significant part of this loss. In addition, reinvesting the saved memory in larger tables can improve performance beyond that of the original full-precision tables while still using less memory. These table memory savings also extend to large existing models: on DeepSeek-V4.1-Flash and Qwen3.8-Flash-Next, low-bit tables save 47-55% of table memory, with perplexity increases of less than 1.2% and largely preserved downstream task performance. These findings establish low-bit precision as a practical way to reduce the memory requirements of n-gram embeddings and use the saved capacity to improve model performance.
memory - arxiv:2610.31644 · cs.CVGigaBench: Benchmarking Learnable Representations for Gigapixel Image Compression and Super-ResolutionMatteo Fusconi, Alex Costanzino, Pierluigi Zama Ramirez, Luigi Di Stefano
Learnable Representations (LRs) can fit parametric families of functions to diverse kinds of signals. These emergent neural-native data formats are gathering interest in the scientific community due to their compactness, resolution-agnosticism and amenability to direct neural processing. On the other hand, gigapixel images are becoming increasingly widespread across a variety of scientific, cultural, and industrial domains. In this paper, we present GigaBench, the first benchmark and experimental evaluation focused on deploying LRs with gigapixel imagery. GigaBench includes images collected, acquired, and synthetically rendered across diverse domains, and addresses image compression and super-resolution. We adapt a representative set of LRs, spanning Implicit, Hybrid, and Explicit methods, so as to operate reliably at the gigapixel scale on consumer hardware. Through systematic evaluation across reconstruction quality, model size, compression and inference times, we identify consistent trends across methods and suggest actionable guidance for future investigation.
benchmark - arxiv:2610.31643 · physics.opticsConfinement of hybridized light modes in nonlinear-optical superpotentialsFederico Presutti, Ryotatsu Yanagimoto
In this work, we present a spatial and temporal light trapping mechanism induced by coherent frequency conversion. We consider a wave-mixing process, where a pumped nonlinearity mediates a frequency-domain beamsplitter interaction between two signal modes. When these wave modes are counterpropagating, temporally trapped eigenmodes emerge at the zero of the nonlinearity profile. We establish that such spatio-temporal trapping has a direct connection to supersymmetric quantum mechanics, which ensures the existence and robustness of the trapped modes and allows us to derive analytic solutions of the bound eigenmodes. Backwards sum-frequency conversion may thus enable a unique class of active photonic memory devices, whose design parameters are compatible with the state of the art in integrated nonlinear photonics. We propose proof-of-principle experiments to study this nonlinear-optical trapping effect on thin-film lithium niobate. In addition, as a case study, we show how the temporal trap can be used alongside parametric gain. Our work unveils potential applications ranging from classical integrated photonics to multimode quantum nonlinear photonics.
memory - arxiv:2610.31641 · cs.CVUniVOE: A Removal-Driven Framework for Interaction-Aware Universal Video Object EditingShuzhou Yang, Lanqing Guo, Lingen Li, Weiqi Li +3
Video object removal and injection are reciprocal editing operations between object-present and object-absent video states, yet they differ substantially in training difficulty. Removal can be effectively bootstrapped from video inpainting and further adapted to erase object-related interaction effects. Injection, however, requires large-scale paired supervision to model object appearance, dynamics, and scene-dependent effects, while perfectly aligned real-world pairs are difficult to collect and synthetic data lacks sufficient realism and diversity. To address this challenge, we present UniVOE, a removal-driven framework for training both video object removal and injection models. We first train an effect-aware remover by transferring video inpainting priors and finetuning on object-present/object-absent pairs. The trained remover converts real object-present videos into object-absent counterparts, creating large-scale paired supervision for injection. An injection model is trained on these pairs in the reverse direction to recover objects together with their dynamics and interaction effects. Moreover, since pixel-level reconstruction alone cannot adequately capture perceptual realism, we introduce a reward-based post-training method that aligns injection results with human preferences using a proprietary reward model. Experiments against strong open-source and commercial systems demonstrate that UniVOE achieves state-of-the-art performance on both object removal and injection, particularly in challenging interaction-effect cases.
post-training - arxiv:2610.31639 · cs.LGCausal Chain of Thought: Mitigating Attention Leakage via Context IsolationGabriele Dominici, Teague McMillan, Fazl Barez, Marc Langheinrich +1
Chain-of-thought (CoT) prompting elicits step-by-step rationales that appear to explain model predictions, but there is no guarantee these steps causally drive the final answer. In standard autoregressive generation, dense attention allows each reasoning step to attend to the full context, enabling the model to derive answers through direct lookup while generating plausible-sounding but post-hoc justifications. We introduce CausalCoT, an inference framework that enforces causal transparency by decomposing problems into a directed acyclic graph of atomic sub-questions, evaluating each node in an isolated context containing only its direct causal parents. This Markovian structure removes attention pathways to non-parent content, making downstream predictions sensitive to local reasoning rather than global context leakage. Across a diverse suite of reasoning benchmarks, CausalCoT preserves competitive predictive performance while improving explanation faithfulness, increasing causal sensitivity of intermediate reasoning steps by an average of 12 and up to 21 percentage points over the strongest baseline. The structured dependency graph additionally enables test-time interventions in which local corrections propagate predictably through downstream reasoning.
benchmark - arxiv:2610.31636 · cs.CLCan an Agent Debug a Web Corpus? Agentic Auditing of Slovak FineWeb2Marek Šuppa, Ivan Vykopal, Jaroslav Kopčan
Multilingual pipelines such as FineWeb2 apply one recipe across a thousand languages, leaving the failure modes of individual languages unanticipated. We use an LLM once, as a corpus debugger rather than a classifier: a tool-using agent explores the corpus through a read-only, budgeted API and compiles each finding into an executable test, which a native speaker then validates. On the 26.5M-document Slovak portion of FineWeb2, three runs costing \$2.21-\$2.84 each found seven findings over four phenomena. The largest is a machine-generated content farm of 1,408,437 documents (5.32%) that passed the pipeline's filters, 86.6% of it passing even stricter thresholds. Blinded labels support 6 of the 7 findings (defect rates .53-1.00 against .08 in a random control) and do not support one. The compiled suite scans the full corpus in 0.7 s of wall-clock time without a GPU, against \$12k-\$90k per LLM classification pass. The best static sampling baseline validates more selectors than any agent run, at under half the cost. The agent contributes exact prevalence, a precision estimate checked against sampled documents, an account of why the pipeline missed each slice, and a reusable test. We release the framework, Slovak test suite, validation data and agent trajectories.
agentagentic - arxiv:2610.31635 · cs.CVHiP3R: Hierarchical Pose Graphs for Long-Horizon 3D ReconstructionZunnan Xu, Shangchen Zhou, Yihang Luo, Chen Change Loy
Feed-forward models accurately estimate camera motion and 3D geometry from short video windows. However, computational and memory costs prevent direct inference on sequences containing thousands of frames. Long videos must therefore be processed as overlapping chunks. This division produces independent local reconstructions with inconsistent coordinate frames and duplicate estimates of shared cameras. Achieving global consistency requires propagating constraints from adjacent overlaps and distant revisits across the sequence. This requirement naturally motivates pose graph optimization. However, the choice of graph level creates a trade-off. Chunk-level graphs efficiently resolve scale drift and close large loops. Yet, their rigid transformations cannot correct internal errors or conflicting frame predictions. Conversely, frame-level optimization offers local flexibility, but scaling it globally is computationally expensive and can break metric scale. HiP3R resolves this trade-off using a hierarchical pose graph. HiP3R first optimizes a robust chunk-level $\mathrm{Sim}(3)$ graph to establish the global layout and scale. It then holds this layout fixed and estimates one shared $\mathrm{SE}(3)$ camera per frame, turning overlapping window predictions into a single trajectory. Experiments across multiple benchmarks demonstrate the effectiveness of HiP3R in improving camera trajectories and 3D reconstruction. Our method applies to both chunk-based and streaming backends without additional training or fine-tuning. The codes are available at https://kkakkkka.github.io/HiP3R.
memorybenchmark - arxiv:2610.31633 · cs.LGBeNice! Closing Looping's Inference-Memory Gap via KV-SharingHayden Prairie, Jacob M. Ryan, Taylor Berg-Kirkpatrick, Daniel Y. Fu
Looped architectures scale quality while keeping parameter count fixed by repeatedly passing activations through the same block of layers. During inference, each recurrence generates independent KV-caches, causing memory to grow linearly. To address this challenge, prior work has proposed KV-sharing (i.e., sharing KV-caches across depth), finding that, in some cases, the KV-cache of the final recurrence can be shared with the earlier recurrences without quality loss, a property we denote as KV-shareability. However, the mechanisms underlying KV-shareability remain poorly understood, with even identically trained looped architectures not consistently exhibiting it. In this paper, we study how KV-shareability can be enabled in looped models. We propose a two-stage method: in the first stage, we use stochastic depth during pretraining, which induces contractive residual dynamics that drive representations toward a common limit across recurrences. In the second stage, we propose BeNice!, a post-training method that broadcasts the final recurrence's KV-cache into the first recurrence, training it to produce compatible residuals while preserving the original model's quality. As a result, BeNice! enables KV-shareability while matching standard finetuning's unshared validation perplexity. Using BeNice!, we show that KV-sharing is complementary to alternative KV-compression methods, composing with memory-efficient sequence mixers (e.g., multi-latent attention, Kimi delta attention) and low-precision KV-cache quantization (e.g., fp8, mxfp8, nvfp4). Finally, in simulated low-memory, long-context inference, KV-sharing improves cache hit rates by up to 31.4%, increases throughput by up to 23.6%, and reduces p90 time-to-first-token by up to 83.6%.
memorylong-contextpost-training - arxiv:2610.31624 · cs.LGQuestioning Instrumental Convergence in RLVR: Checkpoint Evidence from OLMo-3Barış Alparslan, Sandy Tanwisuth, Caleb Petterson
Instrumental convergence is often invoked as a general reason to expect capable agents to acquire or express power-seeking subgoals: self-preservation, resource acquisition, deception, or resistance to oversight which poses an important threat to the society. For LLMs, reinforcement learning is the first stage of post-training that directly optimizes behavior against an objective; therefore, it is the stage where convergent instrumental behavior, on IC grounds, likely to appear. Does reinforcement learning with verifiable rewards (RLVR) make a model more likely to express instrumental-convergence behavior, or does this depend on the training regime, evaluation frame, and stage of post-training? We study this question as a checkpoint-conditioned behavioral trajectory rather than as a single before/after safety score. Using InstrumentalEval, we evaluate OLMo-3-7B-Think across SFT, DPO, 13 intermediate RLVR checkpoints, and the final RLVR model. The trajectory is nonmonotonic: instrumental rate rises from 25.0% at SFT to 46.1% near RLVR step 200, then falls to 11.8% at the final checkpoint. Paired item-level tests and step-wise regression support the final SFT-to-RLVR reduction. Prompt framing does not explain this pattern: a harmful-prompt indicator combined with other framing effects explains only 15% of within-checkpoint score variance on average, and Iterative Null-space Projection (INLP) removes a detectable framing direction without changing the SFT-to-final-RLVR gap. Together with preliminary 32B endpoint results (35.53% SFT to 22.37% RLVR final), these findings indicate that RLVR does not monotonically increase instrumental behavior in this setting: the rate rises early, then falls below the SFT baseline. The result is specific to this model family and benchmark, and judge agreement is moderate, so absolute rates should be read cautiously.
post-trainingbenchmark - arxiv:2610.31623 · cs.CVF2G-Pose: Geometry-Aware Foundation Feature Lifting for Direct RGB-D Category-Level Object Pose EstimationJinyu Zhang, Haitao Lin, Jiashu Hou, Xiangyang Xue +1
This paper studies category-level RGB-D object pose estimation, which recovers an object's 3D rotation and translation without instance-specific CAD models or reference views. This task is highly ambiguous due to intra-class variations, symmetries, and partial observations. Current solutions struggle to balance reliability and efficiency: methods based on intermediate structures are prone to error propagation, while diffusion models suffer from high inference latency and complex post-hoc candidate extraction. To address this, we present F2G-Pose, a real-time direct, single-pass framework that predicts category-level pose, metric size, and camera-frame completed shape from a single RGB-D observation in one feed-forward pass. F2G-Pose lifts dense visual foundation features to partial point clouds, encodes local geometry with DGCNN tokenization, and integrates object-level structure with a Mixture-of-Experts Transformer. Joint size prediction and camera-frame shape completion encourage metric consistency through dense geometric supervision. With its task-specific network trained only on pose-annotated synthetic SOPE data and its frozen RADIO backbone externally pretrained, F2G-Pose achieves state-of-the-art accuracy on SOPE and ROPE while running at 28 FPS for pose-only inference, and further demonstrates strong camera-frame geometry on HouseCat6D and real-object transfer on HANDAL. Analyses of latency, robustness, shape completion, and GenPose++ candidate behavior further demonstrate the efficiency and reliability of F2G-Pose, with real-world manipulation experiments demonstrating deployment potential under segmented RGB-D inputs. The project page, code, and pretrained models are made available at https://jrryzh.github.io/F2G-Pose/.
manipulation - arxiv:2610.31617 · cs.RODiffDreamer: Coupling World Models with Differentiable Simulation for Agile Quadrotor ControlFanxing Li, Yuxiang Huang, Shengyang Wang, Fangyu Sun +6
Tasks such as quadrotor dynamic-gate racing require agents to anticipate external motion while maintaining precise high-frequency control. World-model-based reinforcement learning provides a natural solution by learning latent dynamics for future imagination, yet it often entangles environmental evolution and the robot's own dynamics in a single learned rollout, where model bias can affect policy optimization efficiency. We propose DiffDreamer, a differentiable reinforcement learning framework that tightly couples a learned world model with differentiable simulation for actor optimization. In one optimization step, the learned world model propagates latent states that capture external dynamics and history-dependent features, while the differentiable simulator propagates the robot's explicit physical state. We further decompose the actor objective into a critic-driven pathway for non-differentiable task rewards and a simulator-driven pathway for differentiable motion rewards. Experiments on aggressive tracking and dynamic-gate racing show that DiffDreamer improves training convergence over the model-free, model-based, and differentiable-RL baselines, and final performance over these baselines and classical planning and control methods. Additional experiments assess reward decomposition and the odometry-conditioned relative-state representation through ablations. We also provide proof-of-concept real-world flight experiments for validation.
world modellatent dynamics - arxiv:2610.31613 · cs.ROProgress Probes $P^2$: Lightweight Runtime Monitoring for Vision-Language-Action ModelsAtiksh Bhardwaj, Edward W. Duan, Junwon Seo, Prithwish Dan +3
Vision-Language-Action (VLA) models can perform diverse manipulation tasks directly from visual observations and natural language instructions, but they remain prone to mistakes and unpredictable failures. Deploying these policies without continuous human supervision will require robust runtime monitors that detect when the policy has gone off-course. Existing monitors use external foundation models, such as vision-language models (VLMs) to supervise the policy, introducing significant complexity and computational cost for monitoring at runtime. In this work, we investigate an alternate approach: \textit{can a VLA's own internal activations provide the information needed to monitor its execution?} Our method trains linear probes on policy activations from demonstrations to estimate progress, then flags executions that deviate from expected progress. In experiments with the open-source VLA $π_{0.5}$ and across simulated and real-world manipulation tasks, this approach is competitive with or outperforms VLM-based monitors in task failure detection, as well as methods which require failure data for training. These findings suggest that simple probes of VLA internals can be used to develop reliable runtime monitors, opening a path toward building policies that can act reliably and monitor their execution without external foundation models.
vision-language-actionvlamanipulation - arxiv:2610.31608 · cs.ROHumanoid Mobility via Attention-based ContactsJunzhe He, Kehan Wen, Chong Zhang, Clemens Schwarke +6
Humanoid mobility over complex terrain requires accurate foot placement and coordinated whole-body motion capable of robust adaptation. Reinforcement learning (RL) policies trained without motion references generalize well across terrains, yet eliciting effective whole-body coordination through reward design and exploration alone remains challenging. In contrast, reference-based methods yield natural motions but struggle to adapt beyond the predefined trajectories in their training data. We introduce a learning framework that bridges these two approaches by using motion data strictly as a training prior. An explicit, self-supervised, attention-based contact planner predicts future body-environment contacts, while a contact-conditioned tracker translates these plans into coordinated whole-body motion. Together, they establish contact as an interface between sensory observations and control actions. This abstraction allows the policy to internalize the contact-aware structure of locomotion rather than directly replicating kinematics. Consequently, a single policy can exploit data-driven motion priors for natural whole-body coordination while retaining the flexibility to discover environment-dependent behaviors through reinforcement learning. Building on the learned contact planner and tracker, a subsequent fine-tuning stage bootstraps their acquired capabilities to explore environment variations beyond those covered by the original motion data. Operating entirely from onboard observations without any reference inputs at deployment, this framework enables versatile humanoid mobility in challenging environments.
humanoid - arxiv:2610.31606 · cs.AIHow Ground Truth Shapes Retrieval Scores: Evidence-Aware Evaluation of a Retrieval-Augmented Generation System for Breast LesionsMiguel Mochizuki Silva, Samuel Ismael, Mateus C. Dias, Thaís Gaudêncio +2
Purpose: To describe an evidence-aware retrieval evaluation protocol for a clinical retrieval-augmented generation (RAG) system over breast lesion management literature, addressing the dependence of retrieval scores on annotation coverage. Methods: An 18-document corpus and an 18-query development benchmark, with one Yes/No or WH-question per document, were used. Retrieval combined dense embeddings, BM25, and cross-encoder reranking. Chunk sizes of 128 to 1024 tokens, overlaps of 0 to 20%, and four retriever configurations were evaluated. Relevance was defined by evidence spans mapped to all containing chunks, treating overlapping chunks with equivalent evidence as relevant. Retrieval was evaluated using R-Precision, Recall@K, MRR, and nDCG@K. Separately, generation was assessed on a 512-token index at K=1 to 5 using DeepEval claim-level metrics, with deepseek-v4-pro as the LLM judge, and BLEU/ROUGE metrics. Results: The selected hybrid retriever, using 384-token chunks and 20% overlap, retrieved accepted evidence in the top three results for 17 of 18 queries. Recall@3, R-Precision, MRR@3, and nDCG@3 each reached 0.94. Ground-truth curation increased evidence spans from 25 to 37 and accepted chunk targets from 36 to 51 without changing retrieval rankings. Conclusion: Cross-encoder reranking achieved the highest retrieval scores. Expanded relevance annotations increased recognized evidence without changing rankings, highlighting annotation sensitivity. The small development set and retrieval-informed curation require independent validation, including radiologist assessment of evidence sufficiency.
retrieval-augmentedbenchmarkevaluation protocol - arxiv:2610.31588 · cs.AIVERA: A Responsible AI Dashboard for Assessing LLM Deployments in Light of the EU AI ActRobin Quériaux, Solveig Bachellery, Léa Déléris, Saurabhsingh Rajput +3
Institutions that deploy large language models in regulated settings often need technical evidence of responsible AI practices before approving a deployment. COMPL-AI provides a technical interpretation of the EU AI Act that organises this evidence under six principle groups and eighteen technical requirements, twelve of which current benchmarks can measure. The benchmark suites that produce this evidence are typically operated from a command line, while the staff who approve deployments are non-engineers. We present VERA, an open-source self-hosted dashboard deployed to facilitate responsible AI processes within a large European bank. A reviewer selects a model and runs public benchmarks against it. VERA scores the twelve requirements that current benchmarks can measure and instruments the energy consumed by each evaluation, adding run-level evidence for environmental impact. We run four served open-weight models through the requirement pipeline, and the dashboard displays a score for each measurable requirement. A within-subject study then compares how participants read these benchmark results: from the raw output files in one condition, and through the VERA dashboard in the other. The dashboard reduced the time needed to perform the responsible AI assessment, reducing the time required from 56 seconds using raw files to 30 seconds using our dashboard, on average. Our dashboard scored highly (4.36 out of 5) on usefulness in the survey conducted within the organization. The dashboard and the study material are released publicly to promote replicability and reproducibility in different organisational contexts.
benchmark - arxiv:2610.31573 · cs.AISocial-HUMINT: An LLM-Augmented Unified Framework for Measuring Social Media ExposurePellegrino Casoria, Simon Pietro Romano
In the post-API landscape, social-media profiling has become inexpensive and automated: Large Language Models (LLMs) can extract structured information from unstructured content and deduce attributes users never disclosed. A person's true exposure is not what they post but what can be inferred from it. As these practices become routine, exposure to social engineering attacks becomes an operational certainty, demanding a shared framework for assessment and mitigation. We make this exposure measurable by quantifying how accurately a target can be profiled, and how far that profile enables social engineering attacks. Because no existing model fully spans this continuum, we unify Social Media Intelligence (SOCMINT) and Cyber Human Intelligence (Cyber-HUMINT), two individual-focused disciplines that evolved separately, into a single doctrine-grounded taxonomy of 38 capabilities across the Cyber Threat Intelligence lifecycle, where user profiling bridges passive collection and active engagement. Domain experts validated the taxonomy (Fleiss' $κ$ = 0.92). We operationalize its core path (extraction, inference, and susceptibility assessment) as an LLM-augmented multi-agent pipeline, evaluated on real participants' public data. Susceptibility is assessed indirectly: the pipeline predicts how each target would rank personalized engagement options, matching their actual choices significantly above chance. This ranking capacity captures the personalization on which tailored attacks depend. The underlying profiles are accurate: extraction reaches F1 = 97.3%, and inferred attributes score 4.73/5, both far above rule-based and NER-based baselines at negligible cost. The framework augments security analysts' tradecraft rather than replacing it and enables defenders to gauge their own exposure before an adversary does.
multi-agent - arxiv:2610.31572 · cs.LGOn the Token Collapse Phenomenon in Textual Graph Soft PromptingAndrea Giuseppe Di Francesco, Rishabh Jain, Franco Maria Nardini, Nicola Tonellotto +2
Recent methods for injecting graph structure into LLM prompts include graph soft prompting, in which a graph neural network (GNN) learns graph-specific tokens that capture relational information among entities in textual prompts. G-Retriever combines graph soft prompts with a verbalized description of the same graph to strengthen the textual prompt signal. We conduct an in-depth analysis of G-Retriever and uncover a pathological behavior we term token collapse. Across three benchmarks, three graph encoders, two LLM families, and both frozen and LoRA-adapted models, G-Retriever's graph tokens become almost indistinguishable across inputs. We provide evidence that the same failure is inherited by subsequent methods, even with multi-token pooling. Under the standard setup, swapping a token with the dataset mean, a token from another example, or one derived from a corrupted graph produces no resolved answer-level effect in accuracy. Moreover, a learned constant prompt that never accesses the graph matches or surpasses G-Retriever on all three benchmarks. These findings indicate that, rather than conveying graph content as intended, the collapsed token functions primarily as a generic soft prompt. We also examine conditions linked to this collapse and propose a two-stage procedure that yields non-collapsed tokens while maintaining competitive performance in most cases. However, token variability alone does not prove that the model uses graph information: via interventions and decoding probes, we show that depending on the configuration, a token can reflect the question, an answer prior, or graph-derived content that the LLM may or may not use. Overall, our results offer a practical approach to evaluating graph-token alignment rather than inferring it from task accuracy alone. Code to reproduce our experiments is available at https://github.com/difra100/Token_Collapse_KGQA.
benchmark - arxiv:2610.31569 · cs.ROEcho: Coupled Action and Task Status Signal Denoising for Vision Language Action ModelsYufeng Liu, Bohao Fan, Juntu Zhao, Xinyi Wang +10
Recognizing task completion and failure enables robots to stop upon success and initiate recovery or replanning after failure. Existing methods augment Vision Language Action (VLA) policies with task status signal prediction to support these execution decisions. However, learning and integrating these signals can require fine-grained supervision, external monitoring models, or complex architectural designs. Moreover, the potential of task status signals to directly guide action generation remains underexplored. We present Echo, a framework that combines task status signal learning without frame-level progress annotations with a compact design for status prediction and action generation within a single VLA. Specifically, Echo represents task status with stop and failure signals, added as two dimensions of each action chunk and denoised together with robot actions. Their combination distinguishes ongoing execution, successful completion, and failure. To learn these signals without fine-grained progress annotations, Echo constructs successful, failed, and recovery trajectories with complementary supervision properties. To enable status signal-guided action generation, Echo controls interactions between the signals and actions and denoises the signals faster than the actions. These allow later action denoising steps to condition on cleaner status estimates while preserving accurate signal prediction. Extensive real-world manipulation experiments show that Echo accurately predicts task status and consistently improves policy performance.
vision language actionmanipulation - arxiv:2610.31566 · cs.CVCellTrack-Bench: Microscopic Cell Tracking Benchmark for Vision-Language ModelsIlan Theodoro, Jordão Bragantini, Loïc A. Royer
Large language models have shown outstanding performance on text-first tasks such as coding, math, and writing. Recent efforts are shifting models' training to include scientific problems that often live in highly specialized domains beyond text, such as protein structure and biological images. However, in these scenarios, their general-purpose accuracy compared with specialized algorithms remains in question. One such domain is cell tracking, which requires spatiotemporal awareness of cell dynamics. To evaluate this, we created a new benchmark, CellTrack-Bench, which assesses Vision-Language Models' (VLMs) cell-tracking-related performance against specialized trackers. The full benchmark draws from 15 standard cell-tracking datasets across four imaging modalities and comprises 1,537 problems corresponding to 9,041 cell-linking decisions. We selected a hard subset requiring greater reasoning effort by identifying failures shared across distinct algorithmic trackers. With this proposed dataset, we: (i) evaluated different environment harnesses for VLMs; (ii) benchmarked 9 frontier closed-source models and 13 open-weight models; and (iii) provided a detailed analysis of their failure modes. Code, the stored model outputs, and a pipeline that regenerates the benchmark from the source datasets are available at https://github.com/royerlab/celltrack-bench.
benchmark - arxiv:2610.31565 · cs.CVSquaring-the-Sphere: Distortion-Free and Seamless Panorama Generation via CubeAtlasXuhui Zuo, Chenjie Cao, Tengfei Wang, Junta Wu +2
Generating 360-degree panoramas requires both local perspective fidelity and global spherical continuity. Existing representations face a trade-off: equirectangular projection (ERP) enables seamless panorama modeling but suffers from severe distortion and a domain gap to pre-trained image generators, whereas cubemaps preserve perspective geometry but often produce inconsistent faces and visible seams. We present Squaring-the-Sphere, a panoramic image generation framework built on a novel representation CubeAtlas, which unfolds all cubemap faces into a single compact canvas. By arranging orientation-aligned, rotated polar regions alongside the equatorial faces, this representation preserves the perspective geometry of cubemaps while organizing the cross-face relationships required for seamless panorama assembly. It further enables a pretrained diffusion transformer to model the complete panorama jointly in a single pass, without altering the backbone architecture or positional encoding. While CubeAtlas has a token count comparable to ERP, its structured layout makes many dense attention interactions unnecessary. We therefore introduce a layout-aware sparse attention mask that preserves the interactions required for panorama continuity while removing geometrically irrelevant token pairs. As a result, CubeAtlas achieves better quality than ERP modeling with lower attention computation. Furthermore, we introduce a training-free super-resolution technique and a fusion scheme that enhance fine details in distortion-free perspective space before achieving the final ERP panorama. Experiments on various benchmarks demonstrate that our method improves geometric fidelity, cross-view consistency, and high-resolution visual quality over previous methods.
benchmark - arxiv:2610.31544 · cs.LGSame Luck, Different Choice: Twin-World Credit Assignment for Agentic Reinforcement LearningZeropy Surio, Tobias Engel, Mei-Lin Chou, Anika Raman +1
Agentic reinforcement learning rewards a complete episode, yet the policy needs credit for every decision inside it. Counterfactual branching estimates this credit by re-running the episode after an alternative decision, but the branch draws its sampling noise independently of the factual episode, so the outcome difference mixes the effect of the decision with the luck of everything that follows. We observe that in agentic search the retriever, the corpus, and the grader are deterministic, so an episode is a deterministic function of the policy's decisions and its sampling noise. Twin worlds are therefore realizable: the same noise, a different decision. We introduce TWICE (TWIn-world Credit Estimation), which keys the Gumbel noise of every token by turn and line, rolls out counterfactual twins that inherit the factual episode's downstream noise, and turns their paired outcome differences into turn-level advantages for policy-gradient updates on actor-sampled trajectories; only the actor is deployed. Twin credit is unbiased for any coupling quality, its variance is that of an independent counterfactual scaled by $1-ρ$, where $ρ$ is the twins' outcome correlation, and under coupling the number of rollouts needed to resolve an advantage of size $Δ$ falls from $Θ(Δ^{-2})$ to $Θ(Δ^{-1})$. A square-root rule spends the twin budget where outcomes are predicted to disagree per unit cost. On seven question-answering benchmarks, TWICE improves the average exact match of GRPO by 5.3 and 4.5 points on Qwen2.5-3B and 7B, outperforms tree-search and counterfactual-branching baselines, and uses 68% fewer auxiliary tokens than VinePPO.
agenticbenchmark - arxiv:2610.31538 · cs.LGLACToS: Local Anomaly Detection via Conditional Typicality of SuperpixelsJihoon Hong, Alireza Kheirandish, Rohan Sanda, Sara Fridovich-Keil
Local anomaly detection is the task of identifying spatially confined anomalous or out-of-distribution (OOD) regions within an otherwise in-distribution (ID) image. Most existing methods are designed for a specific context, with a strong focus on industrial defect detection, where they achieve excellent performance on well-established benchmarks. However, we observe that these methods can exhibit significant performance drops in other contexts, especially when the inlier distribution is multimodal. To account for this inlier complexity, we propose LACToS, a principled local OOD detection method based on statistical typicality and made tractable through two innovations: (i) grouping pixels into superpixels and (ii) using a computationally efficient proxy to estimate typicality. We evaluate LACToS across three diverse applications spanning medical imaging, human faces, and X-ray security screening. LACToS performs more robustly than prior methods across these benchmarks. The code is released here: https://github.com/voilalab/LACToS.
benchmark - arxiv:2610.31529 · cs.CLAn Empirical Study of Recent Large Language Models for Knowledge Graph to Text Generation on Standard BenchmarksPriyanka, Renaud Delbru, Souhail Bakkali, Mickaël Coustaty +1
Knowledge graph-to-text (KG-to-text) generation verbalizes a set of knowledge graph triples into fluent natural language. While fine-tuned encoder-decoder models have long been the standard approach, recent LLMs are becoming increasingly competitive. We benchmark recent LLMs against fine-tuned encoder-decoder models, using reference-based metrics and graph-grounded reference-free LLM-based evaluation validated against human annotations. We observed that recent LLMs substantially narrow, and in several settings close, the gap with fine-tuned models on reference-based metrics, while performing better on graph-grounded metrics. LLM-based evaluation scores show limited correlation with reference-based metrics, suggesting that the two approaches capture different aspects of generation quality. Graph-grounded evaluation reveals that prompted models produce fewer contradictions and triple omissions than fine-tuned models, and these errors increase more rapidly with graph size for fine-tuned models, highlighting differences between the two approaches that reference-based metrics may not fully capture.
knowledge graphbenchmark - arxiv:2610.31526 · cs.LGDirect Inference on Admissible Sets (Diona)Lorenzo Sabug, Eric Kerrigan
We propose a new regression method named Direct Inference on Admissible Sets (Diona) that directly uses samples to estimate the state and its derivatives at a query point, along with their admissible bounds, without long global training. By casting the values and derivatives at the query and neighbouring samples as latent variables in a constrained optimisation problem, the burden of long training loops becomes one of on-demand local solves. We establish Diona's theoretical properties including finite-sample coverage and, in the noiseless case, convergence to the true state and derivatives under dense sampling. In a nonlinear regression benchmark on six functions of three to eight dimensions, Diona was more accurate than a Gaussian process (GP) on noiseless data especially with denser data sets. Furthermore, as with GP, it produced uncertainty bounds that contained the true value at more than 95% of query points in four functions. Diona's time to a first prediction after a data set update was three to four orders of magnitude shorter than that of the GP, and two to three orders shorter than that of a neural network, which was more accurate but offers no uncertainty quantification. This low latency is paid for by on-demand predictions, so this latency advantage applies when relatively few predictions are needed between data updates.
benchmark - arxiv:2610.31525 · eess.SYHistory-Aware Control Barrier Functionals with Nonsingular Fading-Memory KernelsEge C. Altunkaya, Esra Demir, Murad Abu-Khalaf, İbrahim Özkol
Safety-critical protection often depends not only on the state and physical limits of the system, but on how that state evolved over time. Here, the safety margin measures the remaining allowance before a physical limit is reached. In this context, the objective is to enforce both physical safety and history-dependent retention of this margin for integer-order control-affine systems. To this end, changes in this margin are converted into memory through fading-memory kernels with finite nonnegative weights. A history-aware control barrier functional (HA-CBFal) is constructed explicitly, with nonnegativity equivalent to retaining a prescribed fraction of this margin's weighted historical value. This condition is enforced by an affine input constraint and proved to imply physical safety under stated kernel, initialization, and feasibility conditions, without finite-dimensional memory realization. In 36 calibration cases on the ADMIRE flight-control benchmark, mean per-case tracking-RMSE reductions of 27.2\% and 7.0\% are obtained relative to calibrated state-only CBF and moving-average control barrier functional (MA-CBFal), respectively, with both constraints preserved to numerical tolerance. In a shorter-recovery maneuver outside calibration with frozen gains, both constraints are preserved to tolerance by HA-CBFal and state-only CBF, but the retention requirement is violated despite MA-CBFal nonnegativity.
memorybenchmark - arxiv:2610.31518 · cs.ROSCOPE: Scene-graph Construction Optimized by Plan Error-correction for Robot Long-term Planning DatasetsChaoyue Li, Zeyu Gao, Yanfeng Lu, Fan Luo
Task and Motion Planning (TAMP) stands as a fundamental challenge in robotics, essential for enabling autonomous agents to execute complex, long-horizon tasks in diverse environments. While Large Language Models (LLMs) introduce a new paradigm for TAMP by leveraging extensive common sense, their inherent reasoning limitations hinder direct application, and fine-tuning strategies to enhance these capabilities are currently bottlenecked by the scarcity of high-quality datasets. Existing automated generation methods often suffer from homogeneity, limiting model generalization, and unsolvability, where hallucinated scene graphs create physical mismatches with task goals. To address these issues, we propose Scene-graph Construction Optimized by Plan Error-correction (SCOPE), an automated framework designed to generate diverse and executable TAMP datasets. SCOPE incorporates multi-level diversity metrics to maximize the information entropy of scenarios. Crucially, it introduces a Plan Feedback Error-correction pipeline that utilizes planner execution failures as feedback signals to retroactively rectify logical inconsistencies in scene graphs, ensuring physical feasibility. Experimental results demonstrate that SCOPE exhibits greater robustness to erroneous scene graphs and generates higher-quality data. When utilized for LLM fine-tuning, SCOPE-generated data yield a downstream TAMP model that achieves an absolute 12.33% gain in success rate over its counterpart fine-tuned with data generated by DELTA, a state-of-the-art PDDL-based planner.
scene graphautonomous agent - arxiv:2610.31511 · cs.AIDuoCook: An Overcooked-Inspired Benchmark for Visual Planning and CoordinationZiyi Jiao, Heng Zhang, Drishti Regmi, Xiaokang Ye +2
Coordinating two players in a cooking task requires a vision-language agent to track changing scene states, sequence recipe steps, and allocate shared resources under a time budget. We introduce an Overcooked-inspired benchmark for evaluating these capabilities in Unreal Engine environments constructed using coding agents. A single vision-language model controls both players from top-down visual observations, selecting high-level actions while the engine executes navigation. The benchmark contains 20 maps in five categories. These categories support evaluation of centralized coordination under equipment contention, changing terrain, and handoff dependencies. A prop legend and pot ingredient-count indicators provide explicit visual cues, while grid-based destinations support spatial grounding. The evaluation compares four VLMs on all 20 maps, measuring task score, order completion, delivery precision, action acceptance rate, and model API expenditure. A memory ablation on the same 20 maps examines the effect of a shared, self-authored notebook while retaining six-response action history in both conditions. The strongest evaluated model, \texttt{gpt-6-astra}, achieves 26.2\% mean order completion. Notebook memory has mixed score effects across models, with positive mean differences for Claude and Qwen and negative differences for GPT and Gemini.
memoryagentbenchmark - arxiv:2610.31495 · cs.LGSABLE: Scalable Affine Bayesian Latent-space ExplorationErik Bodin, Lisa Maria Kreusser, Donney Fan, Henry Moss +2
Complex objects such as engineering designs and proteins are often difficult to optimise because their design spaces exhibit high-dimensional structure and objective evaluations are expensive. Generative models provide continuous latent representations, but their latent spaces can have thousands to tens of thousands of dimensions, creating a regime where the latent dimension $D$ greatly exceeds the evaluation budget. We introduce SABLE, a Bayesian optimisation method for high-dimensional latent spaces with an isotropic distribution and limited budget. SABLE uses a Bayesian affine surrogate for prediction and acquisition, treating higher-order structure as surrogate discrepancy. After $N$ observations, the slope update has dimension at most $N-1$, yielding a decomposition of each query into an in-span component optimised by expected improvement (EI) in at most $N$ dimensions and an off-span component governed by a scalar norm. We show that the off-span norm implicitly selected by affine EI can lead to underexploration, and SABLE instead prescribes it using a finite-horizon schedule, without restricting search to a learned subspace. A recompression scheme reduces information to $\mathcal O(M)$ pseudo-observations, bounding refitting cost by $\mathcal O(M^2D+M^3)$ while keeping parameter inference, prediction, and acquisition decisions to tens of milliseconds. Across synthetic benchmarks and latent-space optimisation tasks, SABLE achieves high sample efficiency and reduces decision time by several orders of magnitude at large budgets.
benchmark - arxiv:2610.31494 · cs.CVAgentVQA: A Unified Benchmark for Agentic Visual UnderstandingAdvait Gupta, Sachin Konan, Tom Yin, Gabriel Sarch +1
Vision-language models (VLMs) can perform a broad range of tasks across diverse settings, yet their performance in agentic contexts remains poorly understood. Although online evaluations in simulators and real-world environments are essential for measuring interactive agent performance, they are costly and challenging to scale. Complementary offline VQA benchmarks can enable faster and more reliable comparisons across agentic domains, yet existing benchmarks remain limited and fragmented. To address this gap, we introduce AgentVQA, an offline benchmark for evaluating VLMs on realistic agentic scenarios across five agent domains: Web Agents, Robotics, Egocentric Videos, Games, and Spatial Planning. We build AgentVQA by manually selecting 1000 challenging, real-world agent examples from 23 source datasets and adapting them for reliable offline evaluation. The resulting benchmark preserves realistic scenarios from interactive environments across diverse domains, and its model rankings track those from corresponding online evaluations. Across the 22 evaluated models, the best achieves only 58.5% accuracy. Our cross-domain analysis shows that errors are driven primarily by failures to ground and maintain accurate representations of the current and evolving visual state, including localizing relevant UI elements, interpreting motion, tracking state changes over time, and inferring 3D spatial relationships, rather than by task misinterpretation or downstream logical reasoning.
agentagenticbenchmark - arxiv:2610.31483 · cs.LGTraining Small Language Models for Transferable Tool SelectionZhaoyi Zhou, Akshayaa Magesh, Jeongyeol Kwon, Artem Kuriksha +3
As LLM agents operate over corpora of more than ten thousand tools, selecting the right tool for a complex multi-step user instruction becomes a central bottleneck. Tool descriptions cannot all fit in context, and relevance-based retrievers only rank tools without knowing which are actually needed. Recent work instead leverages proprietary commercial LLMs like Claude or ChatGPT to drive a retriever, decomposing the instruction and selecting over retrieved candidates. We show that a small, open model can be trained via reinforcement learning to perform the same functionality. On ToolBench, a Qwen3-8B model post-trained with our recipe outperforms Llama-4-Maverick (402B). In addition, the trained model transfers directly to different retrievers and generalizes to new tool corpora in different domains.
llm agent - arxiv:2610.31479 · cs.LGEvaluating the Applicability of LLM Inference Optimizations on Apple SiliconKyrylo Yemets, Andrii Yaroshevych, Victor Muryn, Ostap Khomenko +4
Apple Silicon is among the most widely used platforms for local LLM inference, yet nearly all inference optimizations are designed and validated on server-class NVIDIA GPUs under batched serving workloads. Which optimizations retain their benefits when moved to Apple Silicon, and what determines their applicability? We evaluate four optimization families: KV-cache pruning, speculative decoding, ported GPU kernels, and post-training quantization. Experiments span four Apple machines under a common single-request decode workload, with matched RTX 4090 comparisons where possible. KV-cache pruning can reach up to 1.75 times end-to-end throughput on Apple but can reduce throughput on desktop NVIDIA GPUs. Under the common benchmark, all evaluated speculative decoding methods reduce throughput on Apple, despite reported 2-6 times gains on server GPUs. CUDA kernel designs transfer only partially: public Metal lacks asynchronous copy, and batch-one decode makes several serving-oriented mechanisms inapplicable, although transferable components yield up to 1.20 times decode throughput. Quantization gains vary by machine. Across four families, a reported speedup is not a portable property of a technique alone but a joint function of technique, runtime, hardware contract, and workload regime.
post-trainingbenchmark - arxiv:2610.31478 · cs.LGMoG-VLM: Efficient Video Anomaly Detection through Motion-Guided Vision-Language ModelsKhaled Waleed Dawoud, Muhammad Zaigham Zaheer, Khan Muhammad, Muhammad Haris Khan
Vision-Language Models (VLMs) have shown strong potential for video anomaly detection, but dense frame-level inference makes them impractical for real-time deployment. Existing acceleration commonly relies on uniform sampling, which ignores content dynamics and can either waste computation on static segments or miss localized anomalies. We propose MoG-VLM a motion-adaptive anomaly detection framework that leverages motion purely as a lightweight control signal to decide when VLM inference is necessary. To this end, we introduce Temporal Gradient Magnitude (TGM) which performs content-aware sampling to concentrate computation on informative regions while avoiding redundant processing. We further introduce a hierarchical dual-stream design in which dense multi-frame VLM reasoning is activated only around frames with elevated anomaly likelihood, via anchor-triggered context-window expansion with early termination. A fusion module then reconciles these instantaneous semantic cues with temporal action dynamics. Across four benchmarks, our method achieves 91.91%, 84.30%, 87.88%, and 78.96% AUC-ROC on XD-Violence, UCF-Crime, MSAD, and ShanghaiTech, respectively, establishing new state of the art on three benchmarks under the zero-shot, training-free setting. On XD-Violence, we improve over the previous best result by +0.57% while delivering a 1.31x inference speedup. These results demonstrate that content-aware allocation of VLM capacity, processing fewer frames more intelligently rather than all frames uniformly, is a more effective axis of progress than uniformly densifying VLM inference across frames. Code is available on GitHub: https://github.com/kdawoud91/MoG-VLM.
benchmark - arxiv:2610.31476 · cs.ROPoint-Level Refinable Semantic Mapping for Spatial UnderstandingTaimeng Fu, Yi Du, Zitong Zhan, Qiwei Du +3
Open-vocabulary 3D semantic maps have emerged as a key spatial memory representation for long-horizon robot autonomy, enabling robots to reason about objects, their spatial relationships, and past observations across time. However, existing object-centric semantic maps are primarily designed for static environments and rely on an append-only point integration strategy, where newly observed points are accumulated but previously integrated points are rarely reevaluated. Consequently, perception errors accumulate over time, and outdated points persist after objects move or disappear. Although recent dynamic mapping methods update object-level representations in changing scenes, they do not explicitly revise noisy or obsolete points within each object. We propose an online refinable semantic mapping framework that continuously refines object representations through probabilistic point-level confidence updating. Each object point maintains a confidence score that is updated online using multi-view temporal evidence: consistent observations reinforce reliable points, while inconsistent segmentation masks, depth measurements, and visibility cues suppress spurious and outdated points. Combined with object tracking and re-identification to preserve instance identities under motion and relocation, our framework enables object-centric semantic maps to be continuously refined rather than monotonically accumulated. Experiments demonstrate improved mapping accuracy over the previous method by 3.46% on Replica and 2.55% on ScanNet, show robustness in dynamic environments and under perception noise, and provide an effective spatial memory for object navigation and embodied question-answering with success rates of 82.4% and 76.7%.
embodiedmemory - arxiv:2610.31474 · cs.ROSkeleton-Grounded Latent Action Model for Robot World ModelsQi Lv, Jiayi Lin, Yuxin Jiang, Jianzong Wang +2
Robot world models simulate how environments evolve in response to robot actions. Latent action models enable learning dynamics from videos by encoding visual transitions as proxy actions. However, reconstruction-driven latent actions mix robot motion with object responses and other scene changes, complicating their correspondence with robot controls. We propose SkeLAM, a skeleton-guided latent action model for controllable robot world modeling. SkeLAM decomposes latent actions into a kinematic component grounded by robot skeleton supervision and a residual component retaining complementary visual dynamics. Cross-modal swap reconstruction encourages RGB-derived and skeleton-derived kinematic representations to be interchangeable during video reconstruction. The world model incorporates both components through separate conditioning pathways, while branch dropout reduces its dependence on future-derived residual information. A lightweight adapter maps robot actions to the kinematic space, and subsequent world-model fine-tuning adapts generation to these predicted conditions. At inference, the model generates future observations from the current observation and robot actions. On DROID, SkeLAM reduces end-effector ADE by 19.5% and LPIPS by 11.1% compared with DreamDojo. Motion probing supports kinematic specialization, while qualitative results demonstrate zero-shot human-to-robot motion transfer in generated rollouts. Policy evaluation shows agreement between predicted success rates and those measured in real robot executions.
world modelpolicy evaluation - arxiv:2610.31472 · cs.ROVecTile: A Low-Cost Flexible Tactile Array for 3-DoF Force SensingJian Hou, Xin Zhou, Adam J. Spiers
Tactile sensing provides robots with contact information difficult to obtain from vision alone. Shear force is particularly important for understanding tangential contact, guiding corrective motions, and maintaining grasp stability, yet flexible and low-cost tactile arrays often measure only normal force. We introduce VecTile, a low-cost, 6 mm thin, flexible tactile array capable of three-axis force sensing. VecTile combines a piezoresistive layer with passive silicone nubs that convert normal and shear forces into distinguishable pressure patterns. The sensor can be fabricated in-house without specialized equipment. The complete system costs less than 35 USD for a 4x4 array. We validate the sensing mechanism through finite element analysis and hardware experiments. In controlled single-taxel tests over 1-5 N normal force and up to 1.5 N shear force, VecTile achieves root-mean-square errors of 0.39 N and 0.16 N for normal and shear forces. A VecTile calibrated while flat retains accuracy when curved, while a five-minute manual calibration is sufficient for practical use. Two applications demonstrate its utility. First, shear readings are sensitive enough to stop a sharp scalpel from puncturing a balloon after contact. Second, the sensor substantially improves Diffusion Policy-guided power-plug insertion. All designs and code will be released upon paper acceptance. Project website: https://jiannnh.github.io/VecTile/.
tactilediffusion policygrasp - arxiv:2610.31471 · cs.CVAsk About the Right Moment: Training-Free Object Identity Reasoning for Video-Language ModelsDi Yang, Xiaochang Li, Muyang Liu, Yufei Wang +1
Many questions about a video concern one physical object. Where does the ball end up after the cups are shuffled? Which card is the queen? How often was the mug covered? Video-language models (Video-LMs) recognise such objects in single frames but do not keep them apart over time. On real shell-game videos, open models from 8B to 235B parameters, their thinking variants, a reinforcement-learned tool agent and three training-free tool agents all stay within ten points of chance. We separate semantic identity binding from temporal identity maintenance. An object's identity should be fixed at a moment when the evidence for it is visible, and carried afterwards by a tracker. \emph{Track, Then Bind} implements this principle without training. A promptable video tracker follows every candidate object and draws it with a track number. A frozen vision-language model answers a static question about the identifying evidence, which binds the target to a track. The answer is then derived from the tracked state in text, without asking the model to reinterpret the original video during read-out. A single configuration covers all benchmarks. With Qwen3-VL-32B it solves both VET-Bench tasks, where a fine-tuned trajectory model reaches 92\% and 90\% and the same VLM voting over 16 samples stays at 30--32\%. It answers 55.4\% of the hardest Perception Test cup games, where the best published model answers 41.5\%, and, using track-derived timelines, 48.6\% of TOC-Bench questions, against 37.1\% for the same model answering directly. Drawing the same tracks on the video and asking the question directly falls 12--19 points short of the agent on real cup games.
agentbenchmark - arxiv:2610.31467 · cs.CLFrom Structured Records to Patient Narratives: Preserving Clinical Alignment in LLM-Generated Health SummariesBayram Ayadi, Florian Geissler, Johanna Schmidhuber
As wearable and digital health systems increasingly generate longitudinal physiological records, communicating these observations in patient-accessible language remains a significant challenge. Large Language Models (LLMs) offer a promising mechanism for translating structured clinical information into readable narratives, but this process may introduce semantic drift, where clinically precise meaning is distorted during simplification. We investigate evidence-grounded health summarization for menstrual health monitoring using a constrained generation framework that combines structured evidence conditioning with Retrieval-Augmented Generation (RAG) using a prompt-based constrained generation framework. We evaluate semantic alignment using Natural Language Inference (NLI), atomic factuality metrics, and readability measures. Across three LLMs and 4,500 generated summaries, explicit grounding constraints consistently improve clinical alignment, particularly by improving uncertainty preservation and, once faithfulness is evaluated against retrieved evidence, generally preserving or improving faithfulness, while effects on omission remain model-dependent. We also find that improved clinical alignment comes at the cost of increased reading difficulty and medical terminology. These findings highlight the need to jointly evaluate faithfulness, uncertainty preservation, and readability when developing patient-facing medical LLM systems.
retrieval-augmented - arxiv:2610.31466 · cs.CVAnyGameRenderer: Entity-Consistent Generative Rendering Across Games and an Entity-Level BenchmarkRuicong Liu, Liangyang Ouyang, Fengbo Lan, Zhixiang Wang +1
AI coding agents can now build small playable games in minutes, but the results usually look like programmer art: gray blockouts, colored geometric primitives, and flat semantic blobs. We present AnyGameRenderer, a video-to-video generative renderer that turns such footage into high-quality visuals with characters, motion detail, and effects. Throughout, it keeps the layout, camera, and every entity's position and timing intact. Our AnyGameRenderer generalizes across a wide range of genres that agents build, such as tower defense, auto-battlers, top-down survivors, turn-based battles, kart racing, and crowd runners. AnyGameRenderer takes plain video as its game-state input, with no complex representations such as engine G-buffers, per-object class labels, or low-poly proxies. Gray blockouts, colored primitives, semantic maps, and wireframes all work as input. It also preserves entity correspondence: group counts stay correct, and entities appear and disappear on the right frame. We implement AnyGameRenderer by LoRA fine-tuning MiniMax-H3 ref2va, a 33B reference-video-conditioned video generation model. Training uses $\sim$10k paired clips from three sources: (i) multi-stream rendering data from 38 agent-made games exported from GEAR, an AI game-creation tool; (ii) two-stream rendering data from Unreal Engine, with programmatically controlled crowds, count changes, and entity scales; and (iii) real gameplay recordings, with pairs reconstructed via 3D-box detection. To evaluate the entity correspondence, we propose an entity-level benchmark of 59 held-out cases with 753 entities, scored against ground-truth boxes. Compared with the base model, AnyGameRenderer better preserves entity correspondence: temporal match rises from 50.9\% to 62.9\%, missed count changes fall from 34.7\% to 21.3\%, and fewer entities are hallucinated or left unrendered.
benchmark - arxiv:2610.31455 · cs.CVMixPT: Point Transformer with Interleaved Window and Global AttentionMinkyun Seo, Hyungtae Lim, Jaesik Park
Point cloud transformers aim to capture detailed local geometry and relationships across a scene at a manageable computational cost. Two common approaches are window attention and global attention. They differ in how they enable scene-wide interaction, which refers to information exchange among tokens across a scene. Window attention reduces computational cost by restricting self-attention to groups of tokens called windows; in serialized designs, each window contains a fixed number of consecutive tokens from a spatially ordered sequence. Because each block only directly connects tokens within the same window, information exchange across windows relies on changes in window membership across blocks and hierarchical pooling. Global attention, in contrast, directly connects all tokens within a scene, but computing attention over all token pairs can be costly when the number of tokens is large. To balance computational cost and scene-wide interaction, we introduce MixPT, a point transformer that uses a mix of window and global attention blocks in an interleaved schedule, while retaining a single fixed serialization pattern for its window blocks. For scenes with a large number of tokens, MixPT applies global attention to spatially pooled tokens and returns residual updates to the original-resolution stream. Across indoor and outdoor benchmarks, MixPT achieves competitive accuracy with lower latency than representative window and global attention baselines. We further demonstrate the feasibility of scaling the attention stage in depth by repeating the interleaved pattern of window and global attention blocks.
benchmark - arxiv:2610.31446 · cs.MAWhen Is a Memory Safe Enough to Act On? Action-Conditioned Verification for Long-Lived AgentsYiqi Wang, Yifei Wang, Jiaqi Zhang, Zhangkai Wu +3
Long-lived agents act on memory carried across sessions, and many of their actions are irreversible, so a stale memory can turn directly into harm. Existing work improves the memory itself --- retrieving, refreshing, or reconciling entries --- yet whether an entry must be re-checked before acting is not a property of the entry: the same remembered address can be used unchecked for a routine shipment but should be re-confirmed before a valuable one. We therefore study \emph{action-conditioned memory verification}: given a candidate action, decide which of its memory premises to re-verify and when to proceed or abstain, balancing execution risk, abstention, and verification cost. We formulate this as a sequential stopping problem, in which a verification is worth its cost only if it can change what the agent does next; under suitable verifier reliability and cost conditions, increasing action consequence can move the same memory through proceed, verify, and direct-abstention regimes. Building on this formulation, \textsc{AMV-Guard} is a practical verification gate that extracts the premises a candidate action rests on, calibrates beliefs about whether they are wrong, and selects checks by bounded value of information (VOI). On ToolSandbox, relative to a static policy given the same inputs, it lowers mean verification cost by 33.9\% while keeping 100\% safe-proceed recall and 0\% Gate harmful, and raises Proceed coverage from 38.83\% to 44.17\%; in a gate-core transfer to AppWorld, which carries over the stopping/VOI core but not the full structured pipeline, it lowers verification cost by 21.3\% at a small cost in safe-proceed recall and task success. These results suggest that verification effort should be set by what an agent is about to do, not only by how reliable a memory appears.
action-conditionedmemoryagent - arxiv:2610.31436 · cs.LGConservative Meta-Agent Search: Architecture Adaptation under Non-StationarityAli Atiah Alzahrani
Automated agentic-system design is typically evaluated on static task suites with immediate scalar rewards. Deployed systems face persistent shift, delayed dependent feedback, correlated component failures, and costly architecture changes. We formalize meta-agent architecture search as sequential incumbent replacement with switching costs and hard risk constraints and introduce Conservative Meta-Agent Search (CMAS). CMAS retrieves archived architectures and proposes typed graph edits, then scores each against the incumbent by paired shadow replay under common random numbers, with interventional component credit. Promotion requires an empirically calibrated, variance-penalized simultaneous score above a pre-specified margin, and is followed by a simulated canary stage and pre-specified rollback. Since backtests cannot certify meta-level skill, we introduce MetaInvest-Bench, a regime-switching simulator whose specified worker competence makes a surrogate-optimal architecture computable, with leakage-controlled replay, adversarial governance tracks, and five negative controls. Financial streams serve as a risk-sensitive testbed for meta-level generalization; we claim no alpha. Across 96 streams, CMAS cuts normalized mean surrogate dynamic-oracle regret from 0.116 to 0.104 and raises held-out utility from 0.44 to 0.48 over the strongest adaptive baseline; when extra workers add no value, it contracts to near-single-agent teams. Its deployment inference cost is ~6% lower, though its search-inclusive token cost is the highest of the methods compared. The observed per-update false-promotion rate is 3.4% against a nominal 5% target, and 4.8% in a sealed rerun. The calibration is empirical: it concerns an update-level replay statistic on an author-designed benchmark, and guarantees neither full-utility improvement, nor absence of simulated canary-stage harm, nor sequence-level error control.
agenticbenchmark - arxiv:2610.31432 · cs.ROA Sparse Topometric Graph for Scalable Cross-Session Indoor NavigationEun Sun Lee, Minkyun Seo, Gwangtak Bae, Junho Kim +3
A conventional embodied agent in an indoor environment builds a map by accumulating noisy real-world measurements and results from heavy inference to guide navigation. While the detailed measurements may differ due to scene changes or sensor settings, the unique topological structure of navigable passages or the connectivity between rooms may provide training-free guidance that complements local observations. We propose a sparse topometric graph as an efficient structural representation for robust indoor navigation, alleviating the need for fine-grained matching, which is computationally expensive, memory-intensive, and prone to real-world variations. The map consists of nodes that encode local 3D geometry and distance-weighted edges that capture traversable connectivity. A learned graph-matching network finds consistent subgraph structures across sessions by considering both node-level geometric descriptors and neighborhood connectivity. The holistic contextual matching is robust to local ambiguities caused by weak distinctiveness or repetitive structure, and establishes stable correspondences despite local scene changes, different sensor modalities, or limited spatial overlap. We analyze the practicality of the graph through cross-session point-goal navigation, where a goal specified in a prior reference map must be reached from an unknown initial pose in a later session. Experiments show that our method achieves the highest success rates (77.1% on Gibson and 40.0% under 3D-FRONT scene changes), while reducing memory by more than two orders of magnitude compared to point-level representations. The proposed approach further supports cross-session alignment and goal-directed navigation with strong scalability and robust real-world transfer.
embodiedmemoryagentembodied agent - arxiv:2610.31428 · cs.LGPredictable performance is suboptimal, and what it means for model routingGuanhua Zhang, Moritz Hardt
Large language models (LLMs) differ in capabilities and costs, motivating routing methods that adaptively choose a model for each query. These methods rely on performance prediction: given a query, estimate whether a model will answer it correctly. Most routing methods train separate predictors of candidate models' performance to guide model selection. Our work shows that performance prediction only beats a simple baseline when the candidate model is suboptimal on the task or the performance predictor adds significant capacity relative to the candidate model. Specifically, we take a model's self-confidence, the probability it assigns to its chosen answer, as a baseline and study how much a performance predictor can improve on it. First, we show that any performance predictor that outperforms a model's self-confidence can be used to improve the model's loss: the predictor's advantage transfers exactly in log loss through recalibration. Second, accounting for the predictor's additional parameters, we bound its advantage by the model's suboptimality plus the gain from those parameters. Third, we extend our theoretical results to two-model cascades, which route queries to a more capable model when the first model is predicted to fail. Experiments with 16 language models on four benchmarks show that predictors gain a larger advantage over self-confidence on more suboptimal models. This advantage shrinks as models improve through finetuning. Cascading exhibits the same two patterns: predictor-based routing offers larger gains over confidence-based routing for more suboptimal models, and these gains diminish after finetuning.
benchmark - arxiv:2610.31427 · cs.ROFastHarness: Experience-Driven Review Scheduling for Vision-Language-Action PoliciesTao Lin, Yuxin Du, Haoxiao Wang, Zile Zhong +1
Agentic robot systems combine language-model reasoning with robot policies to monitor execution and recover from failures. However, reviewing every action chunk incurs substantial overhead, even in familiar situations. We introduce FastHarness, an experience-driven execution harness that adaptively schedules GPT-6 Astra supervision of a π0.5 policy. A fast path uses stage memory and reviewed outcomes to reduce review frequency as experience accumulates, while a complementary recovery path focuses supervision on correcting errors. Once recovery is verified, routine execution resumes, with failure experience retained to guide future interventions. The policy remains frozen and generates fresh actions from current observations. Across four simulated manipulation tasks, reviewer token use decreases by an average of 57.2% per task from the first evaluated episode to the mean of the final five episodes. Mean episode time decreases from 85.5 to 52.0 minutes over the same interval. On the same tasks, mean success increases from 80% for GPT-as-Policy (π0.5 + GPT-6 Astra) to 97.19% with FastHarness. These results highlight adaptive supervision as a practical route to more efficient agentic robotics without additional model training.
vision-language-actionmanipulationmemoryagentic - arxiv:2610.31423 · cs.ROFocusVLA: Future-Intent Guided Dynamic View Routing for Vision-Language-Action ModelsHuixin Xu, Ziya Yu, Peng Li
Vision-language-action (VLA) models have become an important paradigm for robotic manipulation by unifying visual perception, language understanding, and action generation. Since reliable manipulation requires sufficient observation of objects, end-effectors, and their spatial relationships, multi-view inputs are widely used to provide complementary spatial information. However, our experiments show that existing VLA models do not fully exploit such multi-view complementarity. The policy tends to persistently rely on the wrist view, lacking the ability to dynamically adjust global-local view dependency across manipulation stages, which limits stage adaptability and robustness. To address this issue, we propose FocusVLA, a stage-aware multi-view routing framework. FocusVLA predicts future interaction intent from historical actions and uses it to guide dynamic view routing during action generation. By adaptively coordinating global and local observations across different task stages, FocusVLA improves multi-view utilization. On RoboTwin 2.0, FocusVLA improves the average success rate over the strongest baseline $π_{0.5}$ by 10.10 pp. In real-world evaluation, FocusVLA improves the success rate from 71.7\% to 85.0\%, achieving a 13.3\% gain over $π_{0.5}$. Moreover, the intent prediction and dynamic routing modules of FocusVLA can be integrated into existing VLA frameworks without modifying their core architectures.
vision-language-actionvlavla modelmanipulationrobotwin - arxiv:2610.31420 · cs.LGPost-Hoc Feature Selection Layer for Neural Networks InterpretabilityJoão Kenji Suwa, João Pedro Silveira e Silva, Bruno Iochins Grisci
The interpretability of complex neural networks remains a critical challenge, especially for models already deployed in high-stakes domains. To address this, we adapted the Feature Selection Layer (FSL) to optimize feature weights post-training, a variant we term post-hoc FSL. Our approach reframes the FSL as a lightweight, trainable module that integrates with already frozen pre-trained models on tabular datasets to highlight the features the original model considers most important. This post-hoc FSL learns feature relevance by fine-tuning its weights based on the original model's learned outputs. Crucially, this process is non-invasive, operating without altering the original model's architecture or its learned parameters. We conducted our experiments using both statistical and visual metrics, including accuracy, F1 score, recall, precision, weighted t-SNE and silhouette score, and also analyzed the stability of the post-hoc FSL on high-dimensional synthetic and real-world tabular datasets. We compared the post-hoc FSL feature weighting method using these metrics against the original embedded FSL and other post-hoc interpretability methods, such as Integrated Gradients, Noise Tunnel, DeepLIFT, Gradient SHAP, and Feature Ablation. Experimental results demonstrate that the post-hoc FSL feature weighting method successfully identified relevant features across the different datasets, maintaining the predictive power of the original neural network while enhancing its interpretability.
post-training - arxiv:2610.31404 · cs.AIAgentPhishing: A Benchmark of LLM Email Agents Under Adaptive AttackJustin Tan, Miao Li, Ding Zhao
Computer-use agents increasingly handle private information and act on users' behalf across desktop and web applications, including email. Phishing therefore poses privacy and security risks that depend not only on whether agents recognize malicious messages, but also on how they respond. Existing phishing datasets were developed primarily for conventional detection, leaving agent susceptibility to persuasive malicious requests insufficiently characterized. To expose vulnerabilities that pre-existing datasets may miss, we introduce AgentPhishing, a benchmark for evaluating whether persuasive phishing attacks can evade detection and induce LLM email agents to comply with malicious requests. We develop PhishOpt, which optimizes reusable email-generation skills against individual agents under fixed benchmark constraints and a realism constraint. We analyze the generated emails using Cialdini's persuasion principles to identify strategies associated with agent failures and use these findings to construct a universal generation skill. Our evaluation shows that agents can handle pre-existing phishing emails safely yet remain susceptible to attacks adapted to agent behavior, while in an AgentPhishing evaluation of nine agents, four exhibit phishing-detection rates below $10\%$ and malicious-request compliance rates above $45\%$. These results highlight the limits of inferring agent safety from conventional phishing evaluations and the need to assess how agents respond to persuasive malicious requests.
agentbenchmark - arxiv:2610.31401 · cs.LGHERD: Harness-Evolving Recursive Distillation for Continual LearningXun Sun, Thanh Thi Nguyen, Thuy-Trang Vu
Continual learning requires a model to keep acquiring new skills and knowledge without compromising its existing capabilities. Parametric methods suffer from catastrophic forgetting and loss of plasticity. Non-parametric methods, such as prompts, skills, and harnesses, leverage in-context learning but are limited by context length and lossy plain-text compression. To effectively combine their complementary strengths, we propose HERD (Harness-Evolving Recursive Distillation), a hybrid continual learning framework that learns efficiently in the harness and internalizes experience into cumulative capabilities via distillation. On the harness side, HERD continually refines its skills to capture reusable patterns, while retaining high-residual examples that the skills fail to capture in a bounded memory. On the parameter side, HERD performs drift-constrained distillation with dual anchors, substantially reducing forgetting over long recursive runs. On the 25-stage task stream of TRACE+17, under fixed memory and compute budgets, HERD matches the strongest parametric baseline in new-task acquisition while exhibiting the least forgetting and the strongest retention of general capabilities.
memory - arxiv:2610.31397 · cs.CLParameter-Efficient Adaptation of Large Language Models to a Low-Resource Agglutinative Language: A Corpus, Datasets, and a Reproducible Benchmark for the Bashkir LanguageMullosharaf K. Arabov, Svetlana S. Khaybullina
Bashkir is a low-resource agglutinative language written in Cyrillic and remains poorly represented in modern LLMs. We present an open resource suite and a systematic benchmark of adaptation methods: the Bashkir Web Corpus (71,567 documents, 46.9M tokens, 16 sources) and three labelled datasets for binary, multiclass and multilabel text classification; the benchmark covers the first two. Experiments use eight architectures (82M-8B: DistilGPT2, GPT-2 base/medium, Phi-3-mini, Mistral-7B, YandexGPT-5-Lite-8B, mBERT, XLM-R) and three adaptation groups (full fine-tuning, QLoRA, in-context learning), plus continual pretraining (CPT). (1) QLoRA on 3.8-8B models outperforms encoder-only fine-tuning on binary classification: YandexGPT-5-Lite-8B with CPT reaches macro-F1 0.748 vs 0.653 (XLM-R) and 0.650 (mBERT); pairwise differences are significant after Holm correction (p<0.001). (2) YandexGPT-5-Lite-8B significantly outperforms Mistral-7B and Phi-3-mini on multiclass classification, BPB and native-speaker ratings (mean 3.28 vs 2.52 and 2.22; n=100; alpha=0.69-0.77; p<0.05); on binary classification its QLoRA advantage is not significant (p>=0.084). (3) CPT effects depend on model and task: perplexity improves for Phi-3-mini (+0.58) and YandexGPT-5-Lite-8B (+0.33) but worsens for Mistral-7B (-1.15, p<0.001); on binary classification CPT helps Mistral-7B most (+4.0 macro-F1). (4) Tokeniser fertility on Bashkir varies 2.67x (2.92-7.80) and tracks adaptation quality, while perplexity across tokenisers is not comparable. Cyrillic tokenisers with Russian pretraining are preferable; LLM judges reproduce native ratings for grammar and coherence (rho=0.33-0.46) but not language correctness (rho=0.16, p=0.11); native evaluation remains necessary. All artefacts are open-licensed.
benchmark - arxiv:2610.31390 · physics.app-phOrbital-coherence memory in electrical transport and its self-consistent Maxwell responsePengfei Zhu
An internal electronic coherence changes electrical transport only when it feeds back on the physical current. We develop a reduced description of this feedback and distinguish its exact consequences from the assumptions required for a microscopic realization. An opposite-parity two-orbital Hamiltonian identifies a current-coupled orbital force. Under an explicitly stated moment closure and collision ansatz, it yields a passive three-state response with an exponentially damped oscillatory memory. The resulting conductivity redistributes a fixed optical weight and predicts a nonmonotonic dependence of dc conduction on coherence damping at fixed remaining parameters. The orbital reduction constrains the coupling and internal frequency through a shared bare splitting, providing a test beyond unconstrained spectral fitting. A charge-conserving spatial extension couples to Maxwell equations through a positive energy balance and separates the constitutive response from longitudinal collective and transverse electromagnetic modes. We formulate complex-conductivity and transmission tests, including the limits of parameter identification over a finite bandwidth. The model supplies a conditional connection between orbital dynamics, transport memory, and electromagnetic observables; conductivity alone does not establish a uniquely quantum mechanism, and a material-specific collision calculation and closure-error assessment remain necessary.
memory - arxiv:2610.31384 · cs.ROCoHelix: Event-Based Multi-Propeller Rotation-Rate Tracking with Shared Projective GeometryRadim Spetlik, Tobiáš Vinklárek, Jan Klivan, Tomáš Kestřánek +2
Estimating propeller rates from a moving event camera requires separating blade rotation from the changing image geometry. Background activity mixes with blade events, while repeated blade appearances admit half- and double-rate ambiguities. We propose CoHelix, a recursive filter that estimates one shared homography and separate rotation states for four rigidly mounted, coplanar propellers. Each admitted event updates its assigned rotation state, and its accepted geometry measurements correct the shared homography used to interpret later events from every propeller. We report lock, the fraction of evaluated time for which an estimate remains within 2% of synchronized ground truth, because mean error alone does not show how long a tracker follows the true rate. With fixed-rate initialization, leave-one-recording-out evaluation on moving-camera data gives CoHelix 62.9% lock, compared with 39.8% for HelixTrack and 22.2% for EventPro. With other settings fixed, allowing four geometry contributors instead of one raises lock from 3.4% to 62.9%. On a private target-follower dataset with a different target quadcopter, CoHelix retains the measurement settings selected on moving-camera data and uses a fixed initial rate. It reaches 95.9% lock, compared with 56.2% for HelixTrack and 61.1% for EventPro.
helixevent camera - arxiv:2610.31376 · cs.AICertified Execution Harness for Agentic AI: Safe and Least-Intrusive Petri-Net Supervisory Control Between Reasoning and ActionChao Gu, Quanyan Zhu
This paper studies the problem of safe tool execution for large language model (LLM) agents under prompt injection. Using Petri nets, we formalize the execution harness between the LLM and its tools as a supervisory control problem, in which marks represent capabilities, approvals and budgets. Since a transition fires only if the marks it needs are present, an injected request whose preconditions are unmet or already spent is refused regardless of its wording. Treating tool outcomes as uncontrollable, we enforce safety by the conventional maximally permissive supervisor, and this safety transfers to the running system under four assumptions. Moreover, to capture an agent that may propose again after a refusal, we introduce the notion of a replanning closed loop, in which each retry costs an inference call. We show that, in this loop, the conventional supervisor can be arbitrarily more intrusive than a stricter safe one. Furthermore, for a given proposer model, the maximally permissive least-intrusive supervisor is characterized as a value-threshold refinement of the conventional one. Under precise conditions, this refinement orders two concurrent actions by at most one monitor place. On an autonomous release agent, we find by synthesis that safe deployment requires a step that the supervisor can refuse after a rollout starts.
agentagentic - arxiv:2610.31372 · cs.AIRIDAR: Safeguarding Research Integrity in Autonomous Research Agents via Automated Detection and Runtime InterventionJiening Siow, Amy Xin, Junjie Wang, Ziyu Tang +6
Autonomous research agents are increasingly used to automate large portions of scientific research, yet they are prone to research integrity violations (RIVs): behaviors that compromise the validity, reproducibility, or faithfulness of the research process. We present RIDAR, a framework for automated RIV detection and runtime intervention. First, we introduce RIVT, a comprehensive taxonomy of 14 RIV types spanning the full research lifecycle. Building on RIVT, we present RIDAR-post-hoc, a framework to detect RIVs within agent trajectories. RIDAR-post-hoc achieves 85.2% F1 performance on RIV-Gold, a benchmark for RIV detection with 60 expert-annotated trajectories. Applying RIDAR-post-hoc to RIV-Wild, a dataset of 300 agent trajectories across 50 distinct research tasks, we detect 3,964 RIV instances in 99.7% of trajectories, revealing the prevalence of RIVs in current agents. Finally, we present RIDAR-runtime, a framework for dynamically detecting and correcting RIVs during agent runtime. RIDAR-runtime effectively reduces agent RIVs remaining at task completion by 36.8%. We will release our code and data to facilitate future research.
agentbenchmark - arxiv:2610.31366 · cs.LGAfterDeleteBench: Auditing Post-Deletion Information Persistence in LLM AgentsEhsan Saleh, Roozbeh Nahavandi, Sama Salarian, Raef Bassily +1
Many LLM agents maintain external persistent memory and derive additional state from past interactions, such as summaries. Deleting an original memory record therefore need not remove information carried by its descendants. We introduce AfterDeleteBench, a controlled behavioral audit for this failure mode in fixed-weight LLM agents. For each synthetic secret, we compare an exposed-then-deleted agent with an otherwise matched agent that was never exposed to the secret, allowing post-deletion recovery to be attributed to the original exposure. In our primary setting, a single query recovers the deleted secret from Qwen2.5-7B-Instruct in roughly two thirds of cases, while matched never-exposed controls have zero recovery. Crucially, recovery does not require a surviving verbatim copy: when the complete secret string is absent and only separately stored components remain under a public composition rule, exact reconstruction succeeds in 62-99% of cases across Qwen, Mistral, and Phi-3. We also show that the response interface is part of the measurement system: target-bearing replies can be scored as failures when the requested format and parser are misaligned. Finally, we separate reconstruction from write-back and fresh-session recall. Provenance-aware closure eliminates observed recovery in the tested positive-baseline settings, whereas an exact-literal write filter blocks reinsertion but cannot erase information already present. Together, these results show that source deletion should be evaluated behaviorally at the level of the agent and its memory system, not only as a storage operation.
memorypersistent memoryagentllm agent - arxiv:2610.31355 · cs.AIMuffin - A Shared Memory Buffer for Asynchronous AI AgentsPing-Lin Kuo, Suman Navaratnarajah, Christian Frey, Gresa Shala +1
The rise of Agentic AI systems has changed the landscape of artificial intelligence by enabling strong reasoning capabilities through specialized multi-agent interaction with structured self-reflection. However, in target-driven reasoning tasks, single- and multi-agent systems are often on par in terms of performance gains. A major challenge is that failures often arise from an ineffective exchange of critical information during inter-agent interaction and coordination. Different from existing multi-agent debate paradigms that rely on synchronous turn-taking and sharing information across different agents, we propose an inter-agent protocol based on shared Memory bUFFer INference - referred to as MUFFIN - in an asynchronous agentic framework. In the first step, an orchestrator routes queries based on their reasoning demands and delegates demanding ones to a multi-agent setting, where a shared memory buffer provides asynchronous exchange of intermediate findings while an agent's private reasoning remains local. Our experiments show that our framework Muffin outperforms established competitors, achieving the best accuracy in 17 out of 20 of our benchmark settings, and provides a better quality-cost trade-off than multi-agent debate: Muffin improves accuracy by 19.2 accuracy points while using 77.4% fewer tokens, without requiring full-trace sharing.
memoryai agentmulti-agentagenticagent systembenchmark - arxiv:2610.31351 · cs.AIAutomating Feature Flag Cleanup at Scale with Multi-Agent LLM Systems: An Industrial Experience at DoorDashAtharv Chandratre, Jai Datar
Feature flags are runtime configuration mechanisms used for gradual feature rollouts, experimentation, and kill switches. Once a feature is fully rolled out, the flag and its gated code become technical debt and can introduce operational risk through inadvertently enabling dormant code paths. At DoorDash, over 56,000 feature flags span 623 repositories, with 1,659 currently stale as of May 2026, with internal estimates of one to two hours per manual cleanup. Existing rule-based cleanup tools proved insufficient in DoorDash's codebases. Our contribution is an end-to-end automated LLM-based feature flag cleanup system that integrates querying live feature flag rollout data using MCP, Jira ticket processing, and end-to-end pull request creation. Our system successfully produced engineer-approved flag removal PRs for 45 (90%) of the 50 stale flags it was evaluated on, proving the system's operational viability at scale with each cleanup averaging only $4.79 and 13.8 minutes. We evaluated the system's quality and reliability boundary based on success rates, cost per cleanup and time metrics. Furthermore, we discuss our lessons learned when architecting this system, to provide practical guidance for developing similar multi-agent systems.
multi-agentagent system - arxiv:2610.31350 · cs.LGGraph-Recurrent Learning of Emergent Collective DynamicsAli Rayat, Acacia Lee, Gia-Wei Chern
Collective ordering in correlated electron systems emerges from microscopic dynamics, yet the evolution of observable textures need not admit a closed, memoryless description. We introduce a graph-recurrent neural network (GRNN) framework that learns such emergent dynamics directly from trajectories without specifying microscopic equations of motion. The framework integrates gated recurrent memory with graph message passing: local hidden states encode information beyond the instantaneous observables, while spatial coupling allows this information to propagate between sites. The observable and hidden fields evolve jointly through a shared local rule, providing a scalable representation of dynamics with spatial interactions and temporal memory. We demonstrate this framework on the emergent dynamics of charge-density-wave order in the semiclassical Holstein model, successfully reproducing temporal and spatial correlations and dynamical scaling during domain coarsening. By learning finite-time evolution directly, the GRNN offers a route to efficient, data-driven descriptions of collective dynamics with unresolved microscopic degrees of freedom.
memory - arxiv:2610.31334 · cs.ROGelSlim 5.0: Unifying Vision and Touch with a Single Fingertip CameraXiangdong Wang, I-Hsuan Huang, Xiaohao Xu, Xili Yi +2
Vision and touch provide complementary cues for object pose and local contact geometry, but external cameras are often occluded at the contact site. We present GelSlim 5.0, a fingertip sensor that acquires visual and tactile observations with one fixed camera and a continuous UV-responsive contact layer. By switching illumination, the camera sequentially acquires through-membrane visual images and contact-sensitive tactile images from a shared viewpoint, avoiding inter-camera extrinsic calibration. To combine these observations, we introduce VisTacFusion, a multimodal transformer with separate prediction paths for planar object pose and contact geometry. Asymmetric bottleneck fusion incorporates visual context while preserving tactile spatial features for depth and surface-normal estimation. Modality dropout enables inference with paired inputs or either modality alone. On real validation data from 12 known objects, paired inputs reduce depth MSE, angular error, and translation $L_1$ by 61.2%, 72.3%, and 61.3%, respectively, relative to tactile-only inference with the same model. Relative to visual-only inference, the corresponding reductions are 93.7%, 6.1%, and 8.6%.
tactile - arxiv:2610.31329 · cs.CVWeVisOne-VL: Unified Visual Perception from Objects to PixelsTianming Liang, Junyi Pan, Yuxi Li, Canmiao Fu +4
Existing vision-language models (VLMs) remain unreliable for fine-grained visual perception. In this report, we present WeVisOne-VL (WeChat Vision All-in-One), a powerful VLM that unifies object-level and pixel-level perception through structured next-token prediction. It supports multiple visual tasks including object detection, referring localization, counting, OCR, segmentation, and matting, enabling localization at any user-specific granularity, from points and boxes to masks and alpha mattes. Architecturally, we integrate a lightweight pixel decoder into the VLM via a special <OBJ> token to generate instance-level pixel outputs. We train the model with a three-stage curriculum SFT followed by mixed-task RL. Across over 30 benchmarks, WeVisOne-VL demonstrates leading performance on both object-level and pixel-level tasks, while maintaining robust rejection for object-absent queries. By releasing WeVisOne-VL to the open-source community, we hope to advance VLMs from broad visual understanding toward truly reliable perception systems for real-world deployment.
benchmark - arxiv:2610.31326 · cs.LGAgenticEvolve: Evolutionary Program Search with Persistent Language Model Agents that Design and Run ExperimentsZe-Yu Peng, Qing-Yu Lan, Zhan-He Wang, Jun Zhang +1
Evolutionary program search with large language models keeps the structure of an evolutionary algorithm and uses the model as its variation operator. Typically, the model generates each candidate program in a single request from parent programs and their scores. We present AgenticEvolve, an evolutionary program search framework in which each island is run by a persistent large language model agent. Between evaluations, the agent designs and runs experiments on the candidate program. To measure the contribution of experimentation, we compare AgenticEvolve with No-analysis, an otherwise identical configuration with experiments disabled. On two heuristic design tasks with three large language models, AgenticEvolve produces better programs than No-analysis and OpenEvolve on both hidden and unseen instances. It reaches the baselines' final quality with far fewer evaluations and comparable cost. On twelve equation discovery problems, it recovers the ground-truth structure of more equations than No-analysis and existing language-model and genetic programming baselines.
agentagentic - arxiv:2610.31325 · cs.MAConsistent but Biased: Chair Personas Trade Noise for Bias in Multi-Agent Clinical DecisionsYitong Bai, Chengyi Zhang, Ziyang Wang
In clinical systems built from several large language model (LLM) agents, one agent often chairs the discussion and makes the final decision. Whether the chair's persona prompt changes that decision, or the agreement between repeated runs that some systems treat as reliability, has not been tested with the case held constant. We ask whether the persona changes the decision or only its timing, how it affects agreement between runs, and which behaviors appear when the prompt describes only personality traits. A simulated department of seven agents discussed 20 published cases of degenerative cervical myelopathy, with the case reports replayed unchanged in every run. Only the prompts of the chair and its deputy changed (no persona, or a consultative, autocratic or avoidant one), written either as rules of conduct or as personality traits. With personas written as rules, the chair chose to operate now in 93% of runs when autocratic and in 2% when avoidant, against 57% without a persona. Surgery was still planned in nearly all decisions, so the personas mainly changed whether to operate now or after more tests. These effects held with a second model and reworded prompts. The autocratic and avoidant personas made the three runs of a case agree more often, but only about as often as chance predicts, and Fleiss $κ$ did not rise. Prompts that described only personality traits kept more than half of the autocratic and avoidant effects on operating now. They reproduced quick decisions, delays and extra tests, but not ignoring red flags, dropping backup plans or passing the decision to the family. Studies should report and vary the chair's persona as they do the model, and correct agreement between runs for chance before calling it reliability.
agentmulti-agent - arxiv:2610.31320 · cs.CVStream4D: Streaming 4D Reconstruction with Training-free Motion Decoupling via Causal KV-Cache AttentionWei Zhang, Ying Feng, Mengtong Wu, Shiqiang Gong +2
4D scene reconstruction requires robust handling of dynamic objects, yet existing streaming methods commonly assume static scenes, leading to pose drift and ghosting artifacts over long sequences. We introduce Stream4D, a training-free, motion-decoupled streaming reconstruction framework that can be directly integrated into causal Transformer backbones (e.g., StreamVGGT, Stream3R). Our key observation is that causal attention implicitly encodes rich temporal motion cues within the KV cache. We adaptively aggregate these cues from multiple attention layers and across temporal windows, and fuse them with depth priors to handle challenging cases such as occlusions and textureless regions, producing precise dynamic masks that decouple static and dynamic regions. We then apply a soft key masking scheme that leverages these dynamic masks to suppress interference from dynamic regions in causal attention, thereby improving camera pose estimation accuracy. Stream4D inherits the efficiency of causal attention and remains compatible with existing streaming architectures. On diverse dynamic-scene benchmarks, Stream4D significantly improves camera pose estimation and static-scene reconstruction quality, yields reliable dynamic masks, and effectively mitigates long-sequence ghosting.
benchmark - arxiv:2610.31317 · cs.AISurveyGraph: Generating Academic Surveys via Methods-Entity Knowledge GraphsYifan Yuan, Yancheng Zhu, Zuoyu Zhang, Xiaogang Peng
Most automated survey generation systems retrieve papers as flat documents, limiting their ability to capture research structure. We present SurveyGraph, a framework built around a Methods-Entity Knowledge Graph (MEKG) encoding method-level follow-on, competition, and shared-resource relations. The graph supports core-paper discovery, retrieval, hybrid outlining, and evidence-gated narrative synthesis. We evaluate retrieval, semantic edges, citation grounding, collection coverage, and generation-stage components. In a controlled seven-method equal-budget benchmark, MEKG-anchored fusion balances recall and structural density, with an independent co-citation check. In a separate archived graph-native diagnostic, graph-aware retrieval nearly doubles recall over flat lexical baselines across eight runs (p = 0.004). A semantic evidence gate achieves AUC = 0.75. Audits of 160 generated sentences find 80.6% anchored-or-partial citation support. Human adjudication of 108 stratified semantic edges finds 68.5% supported-or-partial relations (binary-support Cohen's kappa = 0.64), with improves strongest (80.6%) and extends weakest (52.8%). A SurveyBench-facing collection probe reaches mean reference coverage of 0.521 across four topics with an eight-paper budget. Same-backend Qwen diagnostics on a 30-paper/74-relation graph show reduced organization signals when hybrid outlining is removed. Concurrent-system comparisons provide bounded context rather than matched-backend end-to-end evidence. The results position MEKG as an auditable structural layer for typed diagnostics, relation-specific filtering, and controlled survey generation. Code and artifacts: https://github.com/xzzvsxd/SurveyGraph.
knowledge graphbenchmark - arxiv:2610.31314 · cs.AIWhen LLMs Forget What They Believe: Characterizing State Integrity Collapse in Long-Horizon DialoguesYancheng Zhu, Yifan Yuan
Large language models can repeat an instruction while failing to keep obeying it. We study this recall-compliance dissociation in long-horizon dialogue and define State Integrity Collapse: a model preserves a stated constraint in memory but stops applying it to later turns. We introduce StateTrack-Bench, a controlled suite of conversations covering numerical, format, role/persona, factual-memory, and logical-rule constraints. Across seven provider-pinned model snapshots and five dialogue-depth probes, strict joint retention ranges from 62.4% to 100.0%, with degradation concentrated in symbolic constraints and in one especially brittle model. A forked recall audit shows that 88.6% of observed failures are behavioral-drift failures rather than ordinary forgetting; excluding the brittle outlier preserves this diagnosis (84.4% of failures). Extended evaluations at depths up to 50 turns confirm that the dissociation persists beyond the core benchmark, while Structured State Anchoring recovers up to 24.4 points on immutable constraints. The results show that reliability cannot be inferred from single-turn instruction following alone; evaluation must track whether remembered constraints remain operational as dialogue state grows. Code and artifacts are available at: https://github.com/xzzvsxd/StateTrack.
memorybenchmark - arxiv:2610.31309 · cs.LGExtreme Compression of Per-Layer Embeddings via Cross-Layer Latent Vector QuantizationJiacheng Liu, Yaxin Luo, Xiaohan Zhao, Peng Sun +4
Scaling embedding capacity alongside Transformer computation is an emerging trend in language models. This trend makes high compression rates increasingly useful. Hence, we study extreme compression of Per-Layer Embeddings (PLE), which follow this trend and account for 35.2% of the parameters of Gemma 4 E4B. At such rates, conventional quantizers give each token-layer vector its own code of only a few bytes. We instead develop OPTIC (Output-Path-fitted, Token-shared Indexed Codes), which stores one quantized code tuple per token and decodes it at every layer with a layer-specific affine decoder. Codes and decoders are fitted to the PLE output path rather than to the stored vectors. At about $100\times$ PLE compression of bfloat16 (BF16) E4B, OPTIC retains an aggregate 97.8% of the uncompressed model's performance across seven benchmarks, whereas product quantization, the strongest of six conventional baselines, retains 66.3%. At $594\times$ and $1,002\times$ PLE compression of BF16 E4B, OPTIC retains 90.5% and 84.1%. On the official E4B release with 4-bit linear weights and 16-bit activations (W4A16), OPTIC retains 96.9% at about $100\times$ PLE compression. During inference at about $100\times$ PLE compression, BF16 E4B uses 29.7% less GPU memory than with uncompressed PLE at the same key-value cache size, while running at most 0.40% slower.
memorybenchmark - arxiv:2610.31307 · cs.AISubgoal-aware Memory Graph for GUI AgentDezhao Luo, Daniel Beechey, Derek Yuen, Weilin Luo
GUI agents aim to automate complex user tasks by interacting with mobile interfaces, but ef- fective decision-making in long-horizon set- tings requires reusing prior experience rather than relying solely on the environment ob- servations. Existing memory-based methods typically store experience either at the single- action level or as entire trajectories. Single- step actions are often too fine-grained to carry meaningful semantics, since the success of an individual action does not necessarily imply progress towards the overall goal. In contrast, whole trajectories are often too coarse-grained and difficult to reuse across different tasks. We propose a subgoal-aware memory graph that distils past trajectories into structured, reusable evidence at the level of subgoals. We define a subgoal as consecutive actions shar- ing a single intent, the granularity at which progress toward the task becomes meaning- ful. To reduce redundancy and support cross- task reuse, we organise subgoal experiences as nodes in a graph, where related memories can be merged, extended, and retrieved based on the environment observation. On SPA-Bench, our method improves success rate by 11.11% over the baseline.
memoryagent - arxiv:2610.31306 · cs.AISuperintelligent Software Engineering: Governance That Scales with CapabilityMohammad Mehdi Gholinejad
Software agents are being trained to exceed human engineers; we argue that capability alone does not produce better software engineering. We define superintelligent software engineering as the discipline of building and evolving software with agents whose capability exceeds that of human engineers, in which governance (gates, contracts, decisions and decay control) scales with capability. The argument rests on three claims. First, correctness in a project rests on decisions and contracts that the code does not carry. Second, knowledge decays. Third, once an agent outpaces its reviewers, reading catches a shrinking share of the changes built on stale knowledge. A simple model shows that, past a fixed review capacity, escaping changes grow linearly with throughput while the share that reading catches falls toward zero; governance can lower the slope, capability alone cannot. An exploratory snapshot of 452 of the most-starred public GitHub repositories finds a root-level or .github agent context file in 48.67% (95% confidence interval 44.1 to 53.3); in the median file that names repository paths, 9.5% to 22% of those paths do not resolve, depending on how a reference is read. We then present OSTEON, a discipline for the regime, released with a DOI (10.5281/zenodo.23436020): each project gets a governed knowledge repository of decisions and contracts, every change is checked mechanically against it, and the knowledge itself is checked for decay at every session start; planning is gated, and sessions record what they learn without writing rules. We report no measured effect of OSTEON, mark claims about its mechanisms as opinion and the author's practice heuristics as design rationale, and state the questions that would falsify the thesis.
agent - arxiv:2610.31303 · cs.AIThe Value of Memory in Language AgentsZehao Lin, Erenkai Zhu, Helan Hu, Xixuan Hao +2
Memory enables language agents to carry useful distinctions from past experience into future decisions. Its value depends on which distinctions survive, whether a bounded agent can use them, and what can be recovered or revised as demands evolve. We develop a decision-theoretic framework that follows this value from raw history through persistent state, operational use, and costly recovery and revision. Original and retained value are expressed in units of avoided task loss; risk-cost frontiers describe attainable recovery and preservation across uses. Unknown demand yields value profiles, while interactive feedback changes the distribution of future work. A cost-aware simulation condition and five minimal counterexamples expose the roles of information timing, retrieval access, component interaction, feedback and demand uncertainty. In particular, a pointer with zero mutual information about the answer can still reduce the cost of exact recovery. Diagnostics on LoCoMo, a two-project update task and longitudinal cases exhibit failures of preservation, use and measurement. At fixed retrieved evidence, a saved ledger changes subsequent answers; at fixed agent outputs, an evaluator revision changes scores. The framework connects these distinctions to operational risk bounds and testable principles for allocating memory, preserving recovery options and revising evidence dependencies.
memorypersistent stateagentevaluator - arxiv:2610.31299 · cs.CVPiHand: Latent Hand-Object Interaction Priors for Physics-Based Policy LearningNiklas Berndt, Xianghui Xie, Enes Duran, Yixiao Wang +1
Synthesizing physically grounded dexterous hand-object interactions is important for VR/AR, gaming, and robotics, yet remains challenging because subtle contact dynamics, diverse object geometries, and target trajectories must be jointly respected while maintaining natural grasps. Existing methods either track mocap trajectories with limited transfer to novel objects, or rely on hand-only generative priors that fail to capture coupled hand-object motion, sacrificing contact fidelity and sample efficiency. To tackle these challenges, we introduce PiHand, a Physically grounded Interaction-aware Hand motion synthesis framework. PiHand first trains expert policies that track mocap demonstrations in simulation, distills them into a unified hand-object interaction prior using a goal-conditioned VAE, and finally solve diverse dexterous manipulation tasks through bounded residual exploration in its latent space. With only a single reward and no complex curriculum training, this yields general tabletop manipulation policies that generalize to unseen objects and trajectories on GRAB and DexYCB. The same prior transfers to object catching and bimanual handover, two tasks that are rarely seen in its training data. Across all tasks, PiHand consistently outperforms baselines in all metrics. We will release the processed data and model checkpoints.
manipulationdexterousgrasp - arxiv:2610.31284 · cs.CVRetrieval-Augmented Online Textual Memory for Episodic Memory Video Question AnsweringRaffaele Cali, Giuseppe Lando, Rosario Forte, Antonino Furnari
Long egocentric videos pose a central challenge for multimodal language models: the visual stream must be compressed into a memory representation that can be searched efficiently while still preserving the evidence needed for question answering. Building on an online textual-memory framework in which first-person videos are converted into temporally grounded captions, this work systematically studies the role of retrieval in episodic-memory video question answering. We evaluate how different retrieval units, query formulations, context budgets, and reranking strategies affect multiple-choice answering on QAEgo4D-Closed, isolating retrieval as the main variable while keeping the underlying generated memory fixed. Our best configuration, Parent, matches fine-grained caption subchunks, then provides their full parent captions to the reasoner. It achieves 56.4 % accuracy, matching strong full-memory and retrieval baselines while exposing only a small fraction of the textual memory at question time. More importantly, our analysis shows that better temporal coverage does not necessarily translate into better answers: increasing the retrieval budget improves temporal Hit@K and Recall@K, but can reduce accuracy by introducing visually plausible distractors and uncertain caption evidence. These findings suggest that retrieval-augmented egocentric memory should not be optimized only for evidence recall, but also for the reliability, specificity, and answer-discriminative value of the retrieved context. The study motivates more structured episodic memories that move beyond raw caption retrieval toward evidence-aware representations for long-form egocentric reasoning.
memoryepisodic memoryretrieval-augmented - arxiv:2610.31282 · cs.CVReCoG: Relational Concept Graph for Training-Free PersonalizationChangseon Yu, Hyeokjun Kweon
Training-free personalization adapts a frozen vision-language model (VLM) to recognize user-defined concepts from only a few exemplars by retrieving concept entries at inference time. While recent retriever-based methods enrich each concept with descriptive evidence, they largely treat concepts in isolation, so the retrieved information can be plausible yet non-discriminative, especially when the personalized concept set contains highly confusable, fine-grained categories. We argue that personalization is inherently relational: correct recognition depends on the specific distinctions between a concept and its nearest alternatives. To this end, we propose ReCoG, a Relational Concept Graph for retrieval-augmented personalization. ReCoG represents concepts as nodes and augments them with directed edges that explicitly capture discriminative cues between concept pairs, elicited by a frozen VLM during database construction. At inference, ReCoG retrieves a shortlist and resolves ambiguity via explicit pairwise comparisons conditioned on the corresponding relational edges, selecting the concept that consistently wins against competing candidates. We also introduce FINGER benchmark to test personalized recognition on confusable concept sets. Across FINGER, ReCoG substantially outperforms prior methods, with the largest gains in the most confusable regimes. ReCoG also achieves consistently best performance on established personalization benchmarks including captioning and VQA, confirming that relational evidence is broadly beneficial beyond fine-grained settings. Project Page: https://csy-98.github.io/ReCoG/
retrieval-augmentedbenchmark - arxiv:2610.31280 · cs.LGBeyond Accuracy: Disentangling Decomposition from Typed-State Quality Assurance in Multi-Agent LLM PipelinesMaria Gabriela Alves Zuppardo, Vinicius Cardoso Garcia
Multi-agent LLM pipelines are often credited with gains in software engineering tasks, but it is rarely clear whether a gain stems from the typed contract between agents or from the specialized prompts that decomposition enables. We study this attribution problem in Requirements Engineering, comparing three architectures for classifying requirements as functional or non-functional and then prioritizing them with MoSCoW: a single-call baseline; a two-call variant that reuses the pipeline's specialized prompts but passes an unvalidated intermediate; and a pipeline with a Pydantic-validated typed state. The 18 conditions cover PROMISE NFR+ (625 requirements), three locally hosted small language models, and two languages. The F1 gain over the baseline is associated with the two-call, specialized-prompt variant (up to +0.075), whereas the typed state adds a small difference (-0.003 to +0.013) that is significant in one of six comparisons. Beyond accuracy, the typed pipeline contains out-of-range model outputs at the agent boundary, while the unvalidated variant fails later and drops the requirement, and it shows a small, reproducible increase in inter-model MoSCoW consistency whose cause our design cannot isolate. The study is associative: the contrasts bundle prompts, call count, and staged context, and the results come from three 7--8B models on one dataset. The implementation and replication package are archived on Zenodo (https://doi.org/10.5281/zenodo.23380444).
agentmulti-agent - arxiv:2610.31277 · cs.AICoWriter: Guiding Methodological and Synergetic Human-LLM Co-AuthoringShmuel Amar, Benjamin Lieber, Ori Shapira, Debanjana Kar +3
While LLMs have transformed text generation, standard chat-and-canvas interfaces often reduce human authoring to quite passive supervision, causing review fatigue, over-reliance on AI, and the sidelining of the human role in long-form, source-grounded writing. Prior systems typically address isolated writing phases, leaving comprehensive end-to-end human-LLM collaboration largely unexplored. In this paper, we introduce CoWriter, a proactive agentic scheme that adapts expert human (co-)authoring methodologies from cognitive science to flexibly steer the entire writing workflow while actively encouraging the user to engage. In an exploratory user study, compared to a standard LLM baseline, authors using CoWriter better expressed their perspectives and maintained a higher sense of document ownership, supported quantitatively by $2.2\times$ more successful agent engagements. Furthermore, external reader judges rated the resulting texts higher in "human touch" qualities like storytelling and integrating unique personal experiences. Overall, our results demonstrate that proactively engaging users through established human practices is beneficial for effective LLM-supported co-authoring, providing useful methodologies, techniques and insights for future research on collaborative agents.
agentagentic - arxiv:2610.31276 · cs.AIRIPPLE: Recovery Internalization through Process-Level Policy Learning and Effect-Aware Credit for Long-Horizon Agentic CodingJingxuan Xu, Zihang Xu, Yuechen Luo, Ziqi Miao +8
Long-horizon coding agents can observe a failure and generate a plausible critique yet continue along the same incorrect trajectory. We call this the Critique-to-Action Handoff Gap. We introduce RIPPLE, a recovery-oriented training framework that internalizes process-level guidance into a single self-recovering policy. A prefix-only teacher provides evidence-grounded recovery supervision without access to future actions or terminal outcomes. After lightweight distillation, reinforcement learning assigns stage-specific credit to reflection triggering, diagnosis, and corrective action. Local validity signals gate trajectory advantages so that intermediate recovery decisions do not indiscriminately inherit terminal outcomes. Across four repository-level software-engineering benchmarks, RIPPLE improves the base policy by an arithmetic mean of 4.8 score points, with no auxiliary critic at deployment. Controlled ablations isolate the benefits of local, stage-aligned credit, and matched-prefix interventions show that critique semantics affect subsequent correction. Recovery gains concentrate where the prefix contains sufficient diagnostic evidence and the repository state remains reversible within the remaining budget.
agenticbenchmark - arxiv:2610.31275 · cs.ROLearning Robot Behavior Priors from Non-Robotic Synthetic DataMateusz Odrowaz-Sypniewski, Michael Amir, Jasmine Bayrooti, Ajay Shankar +1
Real robot data, rich across tasks and morphologies, is often treated as the bottleneck for training robot learning models, including embodied foundation models that aim for zero-shot physical transfer. In this paper, we investigate the degree to which such models can acquire necessary structure using data from systems without embodiment. To this end, we test if inexpensive, procedurally generated synthetic data can yield useful priors before a model observes any real-world behavior. We study this through robot action prediction, a natural pretraining objective and a direct test of whether structure learned from arbitrary synthetic tasks transfers to embodied settings. We introduce three data generation pipelines that model mathematical abstractions of robotic control, and use them as pretraining priors for a Transformer. We demonstrate that a model trained purely on this non-robotic data attains non-trivial zero-shot prediction performance on unseen physical robot tasks via an in-context learning mechanism, and that our specialized control priors notably improve upon generic causal priors. We also compare against a compute-matched baseline trained on real robot data, and find that it outperforms our model but only on the datasets it was trained on. These findings suggest that procedurally generated, mathematical abstractions can yield useful robotic priors, offering a potential complement to expensive embodied data collection.
embodied - arxiv:2610.31274 · cs.AIPointerBench: A Diagnostic Benchmark for Fine-Grained GUI GroundingValentino Sacco, Saurabh Pandey, Alexander Florian Jercher, Maximilian Schilling
Multimodal Large Language Models have shown promising performance in computer-use tasks, from planning user workflows to grounding natural-language actions to precise UI targets. Existing end-to-end benchmarks primarily emphasize long-horizon task completion, while GUI grounding benchmarks remain focused on conventional controls and a limited set of professional interfaces. We introduce POINTERBENCH, a diagnostic suite of 1,500 screenshot-instruction pairs probing three complementary capabilities: thin sub-element identification and geometric grounding in spreadsheets (POINTERBENCH-Sheets), character-level grounding in long-text documents (POINTERBENCH-Text), and mixed GUI grounding across 100 professional applications (POINTERBENCH-Pro). We carefully curate each task with pixel-exact target geometry and evaluate both general-purpose multimodal models and 39 specialist GUI grounding configurations. We find text grounding to be the most challenging setting, with Claude Fable 5, GPT-5.6 Sol, and the leading native specialist UI-Venus-2-9B achieving 69.6%, 53.4%, and 25.8% success on POINTERBENCH-Text despite respective 83.2%, 77.7%, and 54.5% average success across POINTERBENCH. These results reveal substantial headroom in fine-grained GUI grounding across frontier and specialist models. We publicly release the benchmark and evaluation code at https://github.com/warmwindOS/pointerbench.
benchmark - arxiv:2610.31272 · cs.AIMemory-Aware Reward Modeling: Learning Human Preferences for LLM Memory UseYijiang River Dong, Zexue He, Cheng Yu, Tiancheng Hu +8
Memory enables language model assistants to maintain continuity across sessions and generate personalized responses, but knowing more about a user does not necessarily produce responses they prefer. Existing evaluations assess memory capabilities without directly establishing whether memory use benefits the user. To study this gap, we collect approximately 1,000 pairwise judgments from 87 participants using their own ChatGPT queries and personal memories. We find that memory-conditioned responses are not consistently preferred over responses generated without memory and in some cases, users strongly prefer the latter. Automatic evaluators also show limited agreement with users and favor memory-conditioned responses substantially more often than users do. Motivated by these gaps, we introduce a memory-aware reward model that learns from user judgments, achieving 72.8\% preference prediction accuracy and outperforming the existing baselines. Our findings motivate evaluation and optimization objectives that align memory use with the preferences of the people whose information is being used.
memoryevaluator - arxiv:2610.31256 · cs.CLIndoNutriQA: An Indonesian Clinical Nutrition Question Answering BenchmarkTirana Noor Fatyanosa, Ummu Ditya Erliana, Mahendra Data, Muhammad Nurwegiono +2
Clinical Question Answering (QA) underpins Clinical Intelligent Decision Support Systems, but progress is severely limited for low-resource languages such as Indonesian by the absence of public domain-specific data. We introduce IndoNutriQA, an expert-curated Indonesian clinical QA benchmark of 911 question-answer pairs built from real clinical-nutrition reports. Rather than a large-scale training corpus, it is positioned as a high-quality evaluation resource: every pair is grounded in a clinically relevant context, enriched with structured metadata (anthropometric assessment, medical diagnosis, document provenance, chunk segmentation). Eight standardised question types following the Nutrition Care Process (NCP), including the Problem, Etiology, Sign and Symptom (PES) diagnosis statement, are answered for every report, which was written by a graduate-level dietetic intern, by one of three nutritionists. Answers are free-form clinical prose rather than extractive spans, and an expert validation study finds 6-10% of reference answers contestable. The dataset supports generative, extractive and retrieval-augmented QA as well as zero- and few-shot LLM evaluation. We report fine-tuned baselines (IndoBERT variants, mT5, RAG) with 5-fold cross-validation, learning curves and bootstrap 95% confidence intervals, and benchmark eight contemporary LLMs zero- and five-shot. We further calibrate our LLM-as-judge rubric against three expert raters (one RD, two nutritionists), finding it human-comparable on factuality and completeness but unreliable on clinical safety, and a human error analysis shows it misses unsupported content that the raters identify. IndoNutriQA is released to foster research in Indonesian clinical NLP and nutritional decision support.
retrieval-augmentedbenchmarkllm-as-judge - arxiv:2610.31253 · cs.CVReinforcing intra-operative tumour localisation with weak supervision from pre-operative appearancesSidhart Sami, Ali Sharjeel, Sri Atluri, Miao Zhao +3
Limited tumour visibility on intra-operative images complicates surgical navigation, including liver hepatocellular carcinoma (HCC) targeting on computed tomography (CT) imaging. This makes labelled datasets difficult to curate for deep learning, and transferring pre-operative delineations through registration remains challenging due to intra-operative deformation and lack of paired pre- and intra-operative data. We propose to transfer tumour appearances learned from pre-operative magnetic resonance (MR) images to guide localisation on intra-operative CT through weak supervision. A classifier is trained to classify tumour presence on pre-operative images. This is applied to windows on intra-operative images, and despite the domain gap, it assigns higher tumour-presence probabilities on average to tumour-containing windows. These weak cross-domain signals guide a reinforcement learning agent to localise tumours on intra-operative images. We compare direct transfer, few-shot adaptation and synthetic transfer (trained using unpaired MR and CT), which translates CT into synthetic MR before localisation. On an external HCC cohort, synthetic transfer outperformed other tested methods and CT-only baselines. Our framework enables cross-modality tumour localisation without paired MR-CT training images.
agent - arxiv:2610.31243 · physics.opticsSecond- and third-order nonlinearities in silicon nitride waveguides via heterogeneous integration of lithium niobate membranesNatale G. Pruiti, Cosmin Suciu, Eugenio Di Gaetano, Jonathan M. Silver +2
Thick, dispersion-engineered silicon nitride is a leading platform for integrated third-order nonlinear optics, supporting supercontinuum and frequency comb generation. However, it lacks the second-order nonlinearity required by many applications. Here we bond pre-fabricated x-cut thin-film lithium niobate membranes onto finished silicon nitride waveguides as a back-end step, compatible with commercially fabricated waveguides. A single silicon nitride waveguide generates an octave-spanning supercontinuum and, through the lithium niobate membrane, concurrent second-harmonic, sum- and difference-frequency generation. These results open new possibilities for thick silicon nitride photonics, from visible light generation to on-chip $f$-$2f$ comb self-referencing, a key step towards miniaturised timing and navigation quantum technologies.
heterogeneous integration - arxiv:2610.31235 · cs.LGMuting Tokens: Exploring Token Sparsification for Efficient Audio TransformersShreya Sajid, Fabrizio Julian Piva, Hiram Rayo Torres Rodriguez, Elif Ceren Gok Yildirim +2
Transformer-based models have shown strong performance across diverse tasks, but their reliance on self-attention mechanisms often leads to high computational costs. Token sparsification, which reduces the number of tokens processed in transformer layers, offers a promising solution to maintain performance while improving efficiency. However, although widely studied in the vision and language domains, token sparsification remains underexplored in audio, with existing work often limited to a single method, model size, or dataset. This work presents a comprehensive evaluation of token sparsification techniques for audio transformers. We assess a range of token pruning and merging strategies across multiple datasets and model sizes, under both post-training and fine-tuning settings. Furthermore, we introduce a novel sparsification approach that combines attention scores with spectral cues derived from the frequency content of token embeddings. Our experiments show that frequency-aware approaches, particularly those based on spectral flatness and low-frequency energy, consistently outperform attention-only methods. We also find that sparsification based on spectrogram patch intensity is especially effective for longer audio inputs. These results highlight the value of incorporating frequency-domain and input-specific information into sparsification strategies to enhance performance in audio tasks.
post-training - arxiv:2610.31228 · cs.CVLightweight Frame Compensation for Real-Time Polyp Segmentation under Frame SkippingWeihao Gao
Real-time polyp segmentation in computer-assisted colonoscopy is hampered by the difficulty of deploying high-accuracy models under strict latency budgets. Frame skipping, which performs inference every K frames and reuses or interpolates the rest, improves throughput but introduces motion blur and lesion misalignment that degrade temporal consistency. To close this gap, we propose the Lightweight Frame Compensator (LFC), a 77K-parameter plug-and-play module inserted before a frozen segmentation backbone. Operating on a three-frame window, LFC fuses consecutive frames via spatio-temporal attention into a single compensated frame, and is trained only on pseudo-labels generated by a differently seeded model, without any expert annotation. Experiments on three polyp video benchmarks show that LFC recovers most of the accuracy degraded by direct frame skipping and, in several settings, reaches the accuracy of frame-by-frame inference, while retaining a 2.05X--2.55X throughput gain. We further evaluate the temporal stability of three backbones spanning CNN, pyramid Transformer, and global-attention Transformer architectures, revealing a spectrum of sensitivity to frame skipping that arises from how locally or globally each architecture aggregates misaligned input. The compensation gain of LFC follows this spectrum: it is largest for high-capacity Transformer backbones, which achieve the best static accuracy but suffer the most under skip-frame inference. LFC thus offers a lightweight, pluggable, and deployment-friendly route to real-time polyp segmentation.
benchmark - arxiv:2610.31222 · cs.CVGLOW-FM: Efficient Generative Low-Light Image Enhancement with One-Step Latent Flow MatchingAriel Lapid, Yuval Haitman, Reuven Peretz, Arnon Netzer +1
Low-light image enhancement (LLIE) aims to recover well-exposed and perceptually faithful images from severely degraded observations. Recent diffusion-based generative methods achieve strong perceptual quality by leveraging powerful image priors, but their iterative sampling incurs substantial computational cost, limiting practical deployment. Despite recent efficiency gains in generative LLIE, high-quality enhancement with a single function evaluation and low overall computation remains challenging, particularly under diverse real-world low-light conditions. We introduce GLOW-FM, a one-step latent flow-matching framework for efficient generative LLIE. GLOW-FM performs conditional transport in the latent space of a frozen Tiny-VAE and leverages Terminal Velocity Matching (TVM) to enable enhancement with a single function evaluation. Matched comparisons with alternative one-step objectives support TVM as an effective design choice, and, to the best of our knowledge, this is its first adaptation to image restoration and specifically to LLIE. Extensive experiments on challenging real-world benchmarks demonstrate that GLOW-FM achieves competitive image quality and state-of-the-art cross-dataset generalization while requiring only a single function evaluation and 34 GFLOPs for the small variant. These findings highlight the potential of GLOW-FM to bring generative LLIE closer to practical deployment through single-step inference and compact latent modeling. Code will be made publicly available upon publication.
benchmark - arxiv:2610.31215 · cs.AITrion-1: Unifying Language, Protein Sequence, and Structure with Diffusion Language ModelsTianyu Lu, Kaicheng Yang, Jiangtao Feng, Keyue Qiu +6
We envision a general-purpose protein foundation model as a unified, interactive model of sequence, structure, and function, capable of understanding, generation, and cross-modal translation while retaining the instruction-following and reasoning capabilities of modern LLMs. Toward this vision, we introduce Trion-1, the first native, unified, multimodal diffusion language model (dLLM) for proteins. Trion-1 is built on two key design choices: (1) a unified discrete token space that jointly represents sequence, structure, and natural language; (2) a dLLM backbone that replaces strict left-to-right factorization with bidirectional or semi-bidirectional token interactions, enabling global residue communication and long-range contact modeling that are naturally suited to structure modeling. This formulation unifies six core capabilities: folding, inverse folding, sequence understanding, structure understanding, sequence design, and structure design, while retaining the native language capabilities of general-purpose LLMs. On established benchmarks, Trion-1 achieves broadly competitive performance. In folding, Trion-1 achieves a TM-score above 0.8 with a single sample under strict homology control, matching or exceeding the existing tri-modal model using 200 samples. For sequence understanding, Trion-1 performs on par with top-tier LLMs such as Claude Opus 4.7 (62.37%). These results support well-aligned sequence-structure-language representations and strong cross-modal generative capabilities.
benchmark - arxiv:2610.31211 · cs.LGEvolving Federated Density Peak Clustering with Shuffler-Based AnonymizationMiha Ožbot, Seiichi Ozawa, Igor Škrjanc
In this paper, we present a novel one-shot federated clustering approach based on evolving fuzzy clustering and shuffler-based anonymization. In federated clustering, each round reveals additional model information to the server and can increase the opportunity to infer properties of local datasets, even when raw samples are never shared. Consequently, fewer client-server exchanges and unlinkability are important, so transmitted information cannot be attributed to any single client. Data-stream clustering algorithms, which learn in a single pass over the data, align well with the one-shot federated learning paradigm, where a single communication round can be sufficient. The proposed Evolving Federated Density Peak Clustering (eFDPC) method combines evolving fuzzy data-stream clustering and density peak clustering in a federated setting. Each client independently evolves micro-clusters from private data mini-batches through cluster addition and merging. To limit communication, density peak clustering is applied to the learned micro-clusters and only the selected peaks are transmitted, rather than the full local models. A shuffler-based layer is introduced between the clients and the server to anonymize contributions prior to reconstruction and global peak selection. We consider a peak-only variant, eFDPC-P, as well as eFDPC-G with server-side label propagation. Experiments on synthetic and real-world benchmark datasets show that eFDPC outperforms recent federated clustering baselines in one-shot settings and scales to higher-dimensional, non-IID data.
benchmark - arxiv:2610.31209 · cs.CVMoving Regions, Not Moving Objects: Segmenting What Actually Moves under Camera MotionWenhua Jin, Tiancheng Chen, Sheng Tang, Jiahao Li +4
Moving object segmentation is conventionally formulated at the sequence and object level, marking an entire instance as moving once it moves anywhere in a video. Recent instantaneous formulations distinguish moving from stationary intervals, but still generate masks at the object level through frozen segmentation foundation models. Because the mask is derived from an instance-level proposal, motion is typically attributed to the entire instance rather than localized to the regions that actually move. Moreover, such instance-level predictions can produce masks even when no part of the instance is actually moving, revealing a more fundamental issue: motion is not an intrinsic property of an instance; it can only be determined by comparing at least two frames. Building on this premise, we introduce MoRSe, which jointly interprets two frames to predict a moving-region mask, distinguishing actual scene motion from apparent motion induced by the camera. A single RGB encoder replaces multi-stage foundation stacks, so no optical flow, point tracks, or separate mask generator is needed at inference. We release MoRSe-5K, a benchmark of 42 sequences with Static, Dynamic-Static, and Dynamic settings designed to evaluate the model's ability to distinguish and localize moving and stationary regions under different camera and scene motion conditions. MoRSe outperforms both specialized moving object segmentation methods and dense 4D reconstruction methods with motion-thresholded predictions, while producing the fewest false positives on sequences where nothing moves.
benchmark - arxiv:2610.31205 · cs.CVA Visual Grounding Benchmark for Retinal OCT with Multimodal Large Language ModelsRonald Fecso, José Morano, Botond Fazekas, Paul Widmann-Sedlnitzky +5
Multimodal large language models (MLLMs) can answer clinical questions about medical images, but trustworthy use requires visual grounding: models should localize the pathology they report rather than rely on spurious shortcuts. Whether MLLMs can visually ground clinical findings in retinal optical coherence tomography (OCT), the most widely used imaging modality in ophthalmology, remains largely unexplored. To address this gap, we introduce RetinaGroundBench, the first benchmark for biomarker visual grounding in retinal OCT. It comprises 815 B-scans exhaustively annotated by ophthalmologists with 4,537 bounding boxes covering 13 biomarkers across 8 diseases. Because MLLMs often fragment a single lesion into multiple partial predictions, we propose AggIoU, a metric that evaluates whether the union of overlapping predicted boxes covers the annotated region. Across 27 open-source MLLMs (0.8-38B parameters), we find that visual grounding remains largely unsolved: the best general-purpose model achieves a macro F1 of just 23.7%, with consistently low recall, despite reaching up to 73% balanced accuracy on biomarker presence questions. To narrow this gap, we developed RetGrounder, an MLLM fine-tuned using an OCT-specific strategy integrating grounding and textual information. Despite its relatively small size (8B), RetGrounder improves grounding F1 to 33.9%, a 43% relative improvement. Our benchmark provides a reference point for measuring progress in the field of OCT diagnosis with MLLMs.
benchmark - arxiv:2610.31198 · cs.AIDifferent Tasks, Different Timesteps: Temporal Task-Aware Quantization for Diffusion Language ModelsOfek Avivi, Eliron Rahimi, Avi Mendelson, Chaim Baskin
Diffusion Language Models (DLMs) introduce unique quantization challenges due to their iterative denoising process. In this work, we reveal that quantization sensitivity is both taskand time-dependent, with different downstream tasks exhibiting distinct precision requirements across denoising steps. We further show that these patterns can be predicted without groundtruth labels or task-specific evaluation metrics. Building on these findings, we introduce T 2Quant, a task and time-aware mixedprecision quantization framework that uses a label-free sensitivity signal to optimally allocate activation precision across denoising steps under a fixed bit-width budget. Extensive experiments across three DLMs and six benchmarks demonstrate that T 2Quant consistently improves the average accuracy of existing quantization methods, achieving gains of up to 6.7 percentage points under the same budget.
benchmark - arxiv:2610.31194 · cs.ROInside-JEPA: From Prediction to Relational Grounding in Latent World ModelsHossein Resani, Javen Qinfeng Shi
Joint-embedding predictive architectures (JEPAs) learn world models in latent space, enabling efficient planning but leaving their states opaque. We introduce Inside-JEPA, which places relational channels for ordered entity pairs inside the JEPA latent state and shows that an inverse-dynamics loss can ground these channels without relation labels when its head reads them, either alone or as part of the full state. On Something-Something V2, latent prediction alone performs no better than a random relation head, while inverse-dynamics supervision that recovers actions from channel changes improves relational AUROC by .065-.103 across ten settings with annotated boxes, by .081 on RH20T robot episodes with held-out tasks, and by .050-.061 across five relations using off-the-shelf detection and tracking. An inverse head on entity latents decodes actions more accurately but barely improves relational grounding; shuffled labels, two label-free objectives, and action labels on only 5% of the pairs also fail to ground the channels. Given an action, predicted relational changes approach the quality of observed future relations (AUROC .599 vs .621). For one-step relational goals, a supervised readout of the 16 channels ranks the correct action among the top five of 65 in 43% of cases, compared with 24% for prediction-only latents and 17% for a frequency prior.
world model - arxiv:2610.31192 · eess.SYLarge Language Models Meet Bayesian Optimization: Towards Self-evolving Wireless Digital TwinsYingzhuo Sun, Haoyun Li, Ming Xiao, Francisco J. Ferrao +1
Wireless digital twins (DTs) are increasingly integrated with artificial intelligence to support online modeling, prediction, and network optimization. However, most existing approaches perform adaptation within predefined learning or optimization structures, which may become unsuitable when wireless tasks and operating conditions change. In this article, we investigate the integration of large language models (LLMs) and Bayesian optimization (BO) to extend wireless DTs from online adaptation towards algorithm-level self-evolution. We introduce an LLM-assisted BO framework in which BO provides a stable and sample-efficient optimization backbone, while the LLM determines whether selected BO components should be retained or adapted based on the current task and accumulated optimization experience. In simulations, we apply the framework to Gaussian process (GP) kernel discovery for a multi-user wireless beam-selection problem. Simulation results show that the LLM-discovered kernel improves the final BO performance by 19.7% on the training tasks compared with the strongest standard kernel baseline. The results demonstrate the potential of combining LLM flexibility with structured BO for self-evolving wireless DTs.
self-evolving - arxiv:2610.31172 · cs.LGPrompting Strategy Effects Depend on the Judge: A Multilingual Study of LLM-as-a-Judge RobustnessAsal Mehrabi, Mohammad Aliannejadi, Clemencia Siro
LLM-as-a-Judge evaluation is increasingly applied to multilingual tasks, yet prompting strategies such as Chain-of-Thought and Role Conditioning are recommended based solely on English benchmarks. Whether their effectiveness transfers across judge models and languages remains unknown: prompt-sensitivity studies are largely English-centric, while existing multilingual judge studies fix the prompt template. In this work, we address this gap by measuring this interaction through a benchmark of 9 prompting strategies, 4 open-weight judges, 27 languages, and 4 tasks across 13 datasets, yielding over 1M pointwise scores on gold-reference outputs, validated against human judgments. We find that prompting strategies are not transferable: the choice of the judge model dominates the choice of prompting strategy, strategy effects reverse direction across judges, and stability degrades in low-resource languages, where these effects are concealed by aggregate reporting. These findings show that prompting effects in multilingual evaluation are conditioned on the judge, task, and language. Consequently, prompting strategies should be validated for the intended evaluation setting, and cross-study comparisons should account for both judge and prompt configuration.
benchmarkjudge model - arxiv:2610.31166 · cs.ROMemory before Goal: Role of Pre-History Memory for Humanoid Loco-ManipulationChenyuan Chen, Ran Ding, Lechen Zhang, Zewen He +1
Humanoid loco-manipulation in dynamic scenes is non-Markovian: cues vanish as objects change or are hidden, or leave the view as the robot turns and walks, yet whole-body vision-language-action (VLA) policies see only the present. We present WholeBodyMem, an object-guided perceptual memory that keeps the pre-history of object events seen before the robot acts on its goal. SAM3 tracks the task objects, and their events causally select key frames within an adaptive budget, which the action expert reads by cross-attention together with object tokens encoding appearance, geometry and time. We also contribute HuMeGo, a humanoid memory benchmark with six simulated and six real-world Unitree G1 tasks. Keeping the right frames rather than more frames, WholeBodyMem surpasses memoryless and uniform-memory baselines, raising its base policy's success from 29.1% to 72.6% in simulation and from 36.7% to 88.3% on the real robot, while also improving the standard SIMPLE tasks (70.2% to 86.9%).
vision-language-actionmanipulationhumanoidmemorybenchmark - arxiv:2610.31165 · cs.AINetwork topology shapes consensus and vulnerability in LLM populationsRoberto Passaro, Ariel Flint, Romualdo Pastor-Satorras, Andrea Baronchelli +1
As large language models (LLMs) are deployed in interacting populations, their collective behaviour depends not only on the models but also on who interacts with whom. Studies of convention formation in LLM populations have mainly tested well-mixed interactions. We place policies extracted from four LLMs on fully connected, random regular, and heterogeneous networks, with $10^3$ to $10^5$ agents playing the naming game. Topology affects how quickly consensus emerges, sometimes whether it emerges at all, and, in rare cases, which convention wins. Sparsest networks slow consensus up to twelvefold. Yet, by preserving correlations between persistent neighbours, they restore consensus in some cases that fail to converge under well-mixed interactions. Networks also govern vulnerability to adversarial attack. We measure the critical mass of committed agents needed to overturn an established convention. Tipping thresholds follow the same ordering across models and conditions: hubs $\leq$ random $\leq$ low-degree nodes. On the most heterogeneous topology, median critical masses are 3%, 16%, and 35%, respectively; targeting hubs requires one-sixth as many committed agents as random placement. For most models, networked populations are easier to tip than well-mixed ones, so well-mixed estimates can overstate robustness. Assessing the safety of multi-agent AI requires mapping who interacts with whom.
multi-agent - arxiv:2610.31163 · cs.LGPrediction Uncertainty Controlled Computer Model CalibrationRyan Heejoon Byun, Dayeon Yoon, Won Chang
Computer-model calibration is usually undertaken in service of prediction, yet neither Bayesian Kennedy-O'Hagan (KOH) calibration nor amortized simulation-based inference is built to control the predictive error of the calibrated simulator. When parameters are weakly identified or the data-model discrepancy is misspecified, the simulator-output bands induced by a nominal posterior level can be far from nominal. We propose a two-layer framework that targets predictive miscoverage directly. First, a conditional flow-matching model learns an amortized posterior for the calibration parameters, trained on simulator outputs perturbed with Gaussian-process discrepancies of varying structure. Second, conformal risk control arranges posterior draws for the target observation into nested candidate sets and selects the smallest one whose fixed-rank pointwise bands attain a prescribed time-averaged miscoverage on a synthetic calibration sample. Under exchangeability of calibration and target observations, the bands control expected miscoverage in finite samples. In a synthetic benchmark, the proposed bands are roughly 20% the width of KOH-MCMC bands with miscoverage near the 10% target. In calibrating an age-structured SEIR model to the 2010 Malawi measles outbreak, they are again sharper and near-nominal, and the weeks they fail to cover coincide with the asymmetric epidemic decline, exposing structural inadequacy of the mechanistic model.
benchmark - arxiv:2610.31162 · cs.CVHiSTEVE 1.0: Harnessing Human-Interface VLA Agents for Open-World GamesZecong Tang, Zian Guan, Yuncheng Guo, Zhanqiu Zhang +2
Building virtual embodied agents that operate in open-ended games like human players is a long-standing goal requiring both long-horizon autonomy and reliable native-interface control. As a procedurally generated open world, Minecraft provides a natural testbed for studying the integration of these capabilities. Existing agents typically provide either native visuomotor control without long-horizon autonomy or high-level planning executed through environment APIs or predefined skills. To bridge this gap, we propose HiSTEVE, a hierarchical VLA agent with a capability-aware agent harness. At the orchestration level, we equip the harness with structured memory for tracking progress and a VLM planner for selecting subtasks within the VLA's capabilities. To reliably determine subtask completion, we further introduce a policy-coupled task completion state detector that conditions on VLA policy context to signal candidate completion. At the control level, a single VLA executes all subtasks with vision-only perception and native keyboard-and-mouse actions. To mitigate interference between heterogeneous GUI and non-GUI interactions, we introduce regime-specific RMSNorm branches and separate mouse-movement token embeddings. We also train the VLA through world knowledge injection, imitation learning, and reinforcement learning with environment feedback to improve reliability and recovery. HiSTEVE achieves state-of-the-art atomic-task performance, scoring 85.0% on MineStudio Bench and a 37.4% average success rate on OpenHA, while also completing long-horizon tasks such as crafting an iron pickaxe from scratch like human players.
vlavla policyembodiedmemoryagentembodied agent - arxiv:2610.31160 · cs.LGTowards Field-Deployable Wearable IMU Systems for Occupational Activity RecognitionIhsan Ul Haq, James Brusey, Farnad Nasirzadeh, Chandan Karmakar
Wearable inertial measurement unit (IMU) systems for construction worker human activity recognition (HAR) routinely report aggregate accuracies above 95 %, rarely under genuine subject-independent (leave-one-subject-out (LOSO)) evaluation, masking the per-individual variance that determines real-world deployment performance. Using the VTT-ConIoT benchmark, we conduct, to our knowledge, the first systematic configuration sensitivity analysis of this dataset: six classifiers across five sensor placements and three window sizes under a single LOSO protocol, a personalised-calibration pilot, and a validated comparison of tabular GAN-based and standard time-series augmentation. Per-subject LOSO accuracy for the main multi-sensor, 16-class configuration ranges from 61.9 % to 88.7 % (mean 77.5 %), a 27 pp spread, though all data are laboratory-collected. Multi-sensor fusion and longer windows both improve accuracy; standard augmentation gives small, consistent gains, while Conditional Tabular GAN (CTGAN) augmentation actively harms accuracy with a demonstrated distributional-fidelity mechanism; and personalised calibration recovers 16.8 pp for the weakest participants from just 30 % of their own data. These findings provide evidence-grounded configuration guidance for scaling occupational body sensor networks towards field deployment
benchmark - arxiv:2610.31158 · cs.CVBeyond Watching and Searching: Benchmarking Agents for Video-centered Deep Research ReportsJiahao Meng, Qize Yang, Yuxuan Wang, Yunfei Chu +3
Deep research agents synthesize open-web evidence into long reports, but current benchmarks do not evaluate research centered on a source video. We introduce Video-Centered Deep Research Report Generation. Given a video and user request, an agent identifies research questions, retrieves external multimodal evidence, and produces an interactive report linked to relevant video moments. This task has direct practical value because it helps users locate relevant moments, verify claims, reproduce demonstrated procedures, and make informed decisions from video content. We present a skill-based workflow and OmniDR-Bench for this task. For evaluation, we propose Omni Atomic Rubric Trees (Omni-ARTs). Each tree contains three dimensions, 23 shared directions, and task-specific atomic binary criteria. The directions test audiovisual grounding, multimodal evidence use, and interactive presentation. Audited Rubric Construction (ARC) builds these trees through independent evidence collection, automated checks, behavioral calibration, and iterative human review. OmniDR-Bench contains 50 Chinese and English tasks across 19 realworld subcategories. It includes 6,414 criteria. Across 18 agent configurations, the best system scores 63.3 out of 100. Current agents still need broader content coverage, stronger causal and logical analysis across modalities, and more accurate citation support. We further analyze models, harnesses, skill configurations, video categories, and research trajectories. The code and data are released at https://github.com/marinero4972/OmniDR-Bench.
agentbenchmark - arxiv:2610.31152 · cs.CVWildFakeToolBench: Benchmarking AI-Generated Image Detectors with Web-Sourced Image ProcessingMingjie Li, Yukun Jiang, Yicong Tan, Yang Zhang
As AI generated images are getting popular these days, discussions on such generations are widely existed in current public forums, software repositories, and photography tutorials. Moreover, with the help of LLM-assisted code translation, malicious actors can adopt such information to evade detectors without specialized knowledge or local GPU execution of the resulting tools. To evaluate such risks, we introduce \textsc{WildFakeToolBench}, a benchmark suite of $156$ executable image-processing tools derived from $220$ public sources across $142$ domains. Each tool retains its source evidence and passes syntax, execution, and output-distortion checks without using feedback from the evaluated detectors. On $105$ generated images from five generators, we evaluate six state-of-the-art detectors with images processed by the tools in our collected bench to assess their effectiveness against such in-the-wild risks. The strongest individual tools achieve attack success rates of $25.25\%$--$100\%$ across detectors, exceeding the strongest of six selected image-processing baselines for each detector. Evaluating subsets of an eight-tool pool yields worst-case coverage above $80\%$ for every detector. Effectiveness is sparse and detector-specific: high coverage across candidates does not imply reliable one-shot evasion. These findings not only highlight the potential of such web-sourced risks, but also show the potential on establishing AI-image detector robustness evaluation framework based on those in-the-wild information.
benchmarkevaluation framework - arxiv:2610.31149 · cs.RONavigating human-robot interaction challenges in private and residential aged care homesDavid Ahmedt-Aristizabal, Mohammad Ali Armin, Akila Pemasiri, Harshala Gammulle +3
This chapter explores how robotic systems and ambient intelligence can enhance care for older adults in residential aged care (RAC) homes and private homes. Although these settings differ, we frame both as extreme and challenging environments for human-robot interaction (HRI), characterised by cluttered and dynamic layouts, diverse resident needs, and limited on-site technical support. We review embodied and non-embodied systems that assist with daily living, safety monitoring, and social engagement, highlighting hybrid deployments where robots complement fixed and wearable sensors. To ensure these technologies are usable and effective, design considerations consider the cognitive, physical, and social needs of older adults. Interaction design prioritises accessibility for older adults with low digital literacy, using robust but simple voice interfaces, small sets of non-verbal cues, accessible screen-based companions, and safe object handovers. Privacy, trust, and acceptability are addressed via practical strategies such as on-device inference, data minimisation, visible consent controls, and transparent data handling aligned with emerging artificial intelligence (AI) and privacy regulation. Evaluation emphasises real-world metrics including false alarms per day, time to recover from faults, staff workload, and long-term system reliability. We close by outlining open research challenges and future directions for scalable, regulation-ready deployments that support aging with dignity, safety, and independence.
embodied - arxiv:2610.31148 · cs.LGPrompt Sensitivity in Tool-Augmented Model ComparisonsMateusz Modrzejewski, Tomasz Radzikowski
Language models are compared on question-answering benchmarks that pose each question in one wording, and a reworded question can change which of two models scores higher. We ask whether giving the models tools makes such comparisons more stable. An accuracy gain cannot answer this, because tools change several steps between the question and the scored answer at once. We propose a matched protocol that separates those steps: the answer readout, reasoning, the decision to call, the tool's answer, and the report supplied without a call. We apply it to six open models in five domains under five wordings. We find that tools raise accuracy where a program answers the question and widen the wording sensitivity of many pairwise comparisons. The widening follows how often a model calls, and call frequency follows the wording for some models and not others. It also depends on the interface and the instructions: a native tool-calling interface and a second tool description leave no detected widening in molecules, and sensitivity persists where tools return intermediate results. These findings show that deterministic reports do not guarantee stable comparisons. We recommend reporting call frequency and comparison stability under several wordings beside accuracy.
benchmark - arxiv:2610.31146 · cs.AIRethinking Fisher Vectors for Second-order Multi-scale Deep Feature Aggregation in Scalable Scene RetrievalAlexy Bhowmick, Sarat Saharia, Shyamanta M. Hazarika
Aggregating CNN responses into a compact discriminative vector, given their lack of geometric invariance, is a challenge. To address this challenge, we propose a codebook-free method to encode second-order information of multi-scale CNN features within the Fisher Kernel framework. We use a scalable retrieval pipeline to show the efficacy of the proposed vector. The increased discriminative power is confirmed by experimental evaluation of our method on two difficult benchmarks (INRIA Holidays, UKBench). The significant improvement in retrieval accuracy over state-of-the-art approaches comes with only a slight dimension increase.
benchmark - arxiv:2610.31135 · cs.LGMixing Matters: Evidence Position Bias Across Sequence Mixers in Long-Context Question AnsweringAman Behera, Namit Solanki, Mehul Anand
Long-context models often use answer-bearing evidence less effectively when it appears away from a prompt boundary. We compare this position bias across Transformer, state-space, and hybrid Mamba architectures by moving one gold document through a fixed set of nine distractors. The evaluated Transformers show a beginning advantage that the evaluated state-space models lack, while both families favor evidence near the end. A matched pure-versus-hybrid Mamba comparison points in the same direction but remains statistically uncertain. The family difference appears among models that can answer the task. Transformer primacy reappears on synthetic retrieval, where state-space ceiling prevents a family comparison. A corpus swap within one state-space architecture changes overall accuracy without a detectable change in curve shape. Prompt changes can create or nearly remove a beginning advantage. Position remains decodable across architectures even when response curves differ, and attention concentration near initial tokens covaries with beginning sensitivity in the tested Transformer family. These exploratory results show that effective context use depends on the evaluated architecture family, capability, task, and prompt, without isolating the sequence mixer as the sole cause.
long-context - arxiv:2610.31133 · cs.LGIs High Local Accuracy Enough? Geometric Consensus with Support-Adaptive Statistical Personalization for Federated LearningJunyu Lin, Bosong Qi, Pengcheng Shi, Jiayuan Li
Real-world deployment of personalized federated learning (PFL) requires adapting to heterogeneous clients while preserving recognition performance beyond local distributions. However, high local accuracy can conceal failures on locally rare or absent classes. Effective personalization should therefore strengthen global recognition while raising the personalization floor. We propose an evaluation protocol that complements local accuracy with global, rare-class, and unseen-class evaluations to reveal generalization obscured by local scores, and assesses personalization floors among the worst-performing clients to expose weaknesses hidden by aggregate performance. To meet these deployment challenges, we propose FedGRC, a unified PFL framework integrating generalization and personalization through (i) Consensus-Guided Representation Learning (CGRL), which learns discriminative representations through shared directional and magnitude guidance; (ii) Multilayer Metric Synthesis (MMS), which constructs a shared covariance metric from these multilayer representations for statistical classification; and (iii) Support-Adaptive Shrinkage Personalization (SASP), which adapts class means within this metric by shrinking local estimates toward shared means according to local sample support, enabling adaptation while retaining shared knowledge. Extensive experiments, including benchmarking against 21 methods across 5 heterogeneity settings on 3 datasets and comprehensive ablation studies, validate the effectiveness of FedGRC. On CIFAR-10 and CIFAR-100, FedGRC surpasses state-of-the-art (SOTA) baselines in global recognition and personalization floors by at least 2.17% and 2.13%, respectively. On the real-world heterogeneous FedRS dataset, it outperforms the respective best baselines by 2.15%, 3.30%, and 4.61% in local adaptation, global recognition, and the personalization floor.
benchmarkevaluation protocol - arxiv:2610.31130 · cs.LGPost-Audit Mirage: Identifiability Limits of Offline Agent VerificationAman Behera, Payal Payal
Offline tests can fail to reveal whether an agent update will remain safe after deployment. We operationalize this post-audit mirage using matched safe and harmful lifecycles that expose identical verifier-visible evidence while an isolated scorer records opposite deployment truth. Our theorem applies standard observational-equivalence and indistinguishability ideas to this executable audit boundary and states the resulting error-abstention tradeoff. The Post-Audit Mirage benchmark makes that boundary executable across authorization, scheduling, and shared-capacity environments. IdentifiedRangeMonitor uses fresh group-labelled harm outcomes to return deploy, hold, cannot determine, or unsupported. Experiments show that monitoring can recover the missing distinction in favorable settings but may remain undecided for rare groups, small effects, or missing observations. Verifier-agent evaluation must therefore test both calculation validity and whether the available evidence supports the deployment decision.
agentbenchmark - arxiv:2610.31125 · cs.LGDelta N-gram: Forward-Only Continual Writing into a Language Model's Native N-gram MemoryHao Wu
Our previous analysis found that a model's native hashed N-gram table is where continual updates disturb the base least, but its writes needed backpropagation and changed the pretrained table. We present Delta N-gram, a forward-only write path into that table. The pretrained table T^0 is never modified. Evidence is written into a sparse patch Δon the rows it addresses: a 4.1M-parameter writer emits a target and a gate per (position, hash head) event, and a block-averaged delta rule keeps each row a convex combination of its old value and the gated targets. The writer is trained offline on raw text with the pretraining objective; at use time writing takes forward passes only, and clearing Δrestores the original model. A 0.6B prototype trained to answer questions keeps 16/16 facts after a second write; online gradient writes keep 2--4/16. We study two writers: a plain writer that encodes evidence with the T^0-only model, and a delta writer that reads its current memory while encoding. On Qwen3.8-Flash-Next, the plain writer's memory is document-specific: on 64 held-out news events it lowers continuation NLL by 0.068 nats/token against empty memory and by 0.024 against another document's memory, and 0.063 and 0.013 after 16 intervening articles. The delta writer's memory is more document-specific and keeps more of it under distraction (70% vs 54%). Over 35,000 writes on our previous seven-task chain both writers stay stable and the delta writer drifts less (layer-2 math drift 176% vs 188%); later accuracy changes by -0.26/-0.52pp (-7.29pp for gradient updates) and small-k_i drift is 17.9/17.5% (39.2%). Acquisition is negative, and general-input drift reaches 32.8% (20.75%): format rows read by every example grow to 4--12 times their pretrained norm. Zeroing eight such rows removes over 98% of layer-2 drift on math inputs, together with their NLL gain.
memory - arxiv:2610.31123 · cs.CVWildDepth-v2: A Synchronized RGB--LiDAR--Radar Dataset and Benchmark for Metric Depth, Motion, and Tracking of WildlifeMuhammad Aamir, Sangyun Shin, Naoya Muramatsu, Matthew Wijers +3
Measuring animal distance and motion in metric units underpins abundance estimation, morphometrics, and movement ecology, yet monocular video lacks metric scale and wildlife datasets lack metric ground truth. We present WildDepth 2, extending the synchronized RGB-LiDAR WildDepth dataset in three directions. (i) Imaging radar: 37 RGB-LiDAR-4D-radar sequences (19,716 frames) with per-point Doppler velocity and a measured error budget, to our knowledge the first imaging-radar wildlife recordings. (ii) Benchmarking: a frozen, checksummed hidden test set for RGB-only distance estimation and RGB+sparse-LiDAR depth completion, evaluated separately on animal and background pixels. (iii) Motion and tracking: metric 3D trajectories obtained by lifting 2D tracks using LiDAR, radar, or monocular depth. Across 48,723 animal frames, object-level radar calibration improves monocular depth over scene-level LiDAR calibration without requiring LiDAR at inference (AbsRel 0.120 vs. 0.198), although animal-level LiDAR remains more accurate (0.052). Across 431 tracks, once-per-track radar scaling reduces 3D position error four-fold (2.20 to 0.58 m) and distance-sampling detection-function bias from -13% to -2%, matching once-per-track LiDAR calibration. Transferring radar Doppler onto LiDAR produces motion masks 3.7 times denser than radar alone at 72% precision, reduces moving-animal LiDAR accumulation error by 37%, and estimates radial speed (r = 0.60), albeit with approximately 50% underestimation.
benchmark - arxiv:2610.31121 · cs.RORGraph-Eval: Task-Grounded Scene Graphs for Evaluating Robot Video GenerationAleksandr Lazarev, Sofia Zimina, Roman Kuznetsov, Tatiana Zemskova +2
Generated robot videos can look convincing while depicting the wrong interac- tion: a different object is moved, a grasp disappears, or a placement ends out- side its destination. We introduce RGraph-Eval, a reference-based framework that makes these discrepancies explicit through task-grounded temporal scene graphs. Unlike approaches that construct separate evidence representations for different evaluation dimensions, we organize task entities, interactions, and motion within a unified, interpretable scene graph representation. A foundation-model pipeline identifies manipulation roles, grounds persistent object masks, infers and verifies state transitions and action intervals, and reconstructs manipulated- object motion in three dimensions. Applying this representation to a generated video and a reference demonstration supports an interpretable RGraph-Eval score with visible-object, relation-change, and 3D path-shape components. We also introduce ShareRobot-SG, a collection of 1,000 manipulation scenes from ten ShareRobot sources annotated with the multimodal representation. Our eval- uation separates graph-generation quality from agreement with human video judgments, and includes comparisons with adapted SVG2 and SG-Ego, a bench- mark of three video generators. The framework connects robot-video assessment to the observable entities and interactions underlying a task, while making the graph annotations reusable beyond evaluation. Code will be publicly available at https://github.com/SashaDance/rgraph-eval.
manipulationgraspscene graph - arxiv:2610.31118 · cs.AIProductWebBench: Evaluating Website Continuity Beyond Task CompletionJiajun Wu, Jian Yang, Yaxin Du, Shukai Liu +8
Frontend coding agents are usually graded on one question: did the requested target get built. Real engineering changes carry a second obligation, because the rest of the site has to keep working. We call that second obligation website continuity, and introduce ProductWebBench, a benchmark that scores both on living frontend repositories: 320 Change tasks over existing code and 80 long-horizon Build tasks. Every task freezes a repository snapshot, staged requirements, browser evidence and deterministic checks before evaluation, and enters only once a reference passes every check and plausible-wrong ones fail. A simulated user reveals the requirements one stage at a time, and every check cleared earlier stays armed for the rest of the run. Across 8,737 runs from 13 models the two come apart: of the Change runs passing every requested-work check, 28.2% still fail a continuity check -- a blank route, horizontal overflow, console errors -- and 10.8% push a browser state they had already passed back to failing. Because the same checks also run on the untouched repository, each regression carries exact provenance, not a post-hoc label: on the 586 runs where both readings exist, 18.3% are caught breaking a verified state. Interaction is a further blind spot, failing in 29.5% of runs whose static DOM already passes. ProductWebBench therefore scores a patch against the living system it lands in, not the target alone.
benchmark - arxiv:2610.31117 · cs.CVThrough a PRISM, Clearly: Step-Sequence Reasoning for Mistake Recognition in Procedural VideosOlga Loginova, Oleksandr Bezrukov, Frank Keller
Procedural mistake recognition requires identifying deviations from valid multi-step activities. We study post-completion recognition from videos with benchmark-provided step boundaries. We introduce PRISM (Procedural mistake Recognition via Interpretable Semantic Modeling), which grounds video segments in language and maps their descriptions to procedural states using a lightweight hidden Markov model trained on correct executions. Reference alignment and semantic checks identify errors in step content and order. To support evaluation across heterogeneous benchmarks, we map dataset-specific annotations to four edit-based error types: insertion, deletion, substitution, and transposition. Across three benchmarks, we compare gold and VLM-generated descriptions in raw, semantic-role, and action-object forms. Decoder controls show consistent gains from the trained local decoder over nearest-candidate matching, with additional, dataset-dependent benefits from sequence decoding. Semantic role labeling improves Any-F1 in all six matched grounded comparisons, although the gains vary across datasets and do not translate into uniform improvements under separate configuration selection. These results establish language grounding as an effective interface for explicit procedural mistake reasoning and characterize the contributions of representation and decoding.
benchmark - arxiv:2610.31113 · cs.AICan AI Co-Design Structures? An AI Agent for Early-Stage Structural DesignRicardo Maia Avelino, Mennatallah El-Assady, Tom Van Mele, Philippe Block
Recent advances in large language models and agentic workflows offer a new paradigm to early-stage structural design translating design intent into structural models. Although text-to-3D approaches have demonstrated that natural language can be used to produce visual, mesh-based geometric models, structural design requires domain-specific geometric arrangements and semantically rich data types. This paper introduces a proof-of-concept workflow for prompt-to-structure generation using an AI agent connected to a COMPAS-based structural modelling API through a Model Context Protocol (MCP) server. Through controlled modelling operations, the agent creates and modifies a graph-based structural data model composed of explicit building elements. The resulting artifacts can be inspected, modified, serialised, and further connected to downstream AEC workflows. The approach is demonstrated through selected prompt-to-structure examples and by progressive and editable design sequences that resemble typical iterative structural design workflows. The results are preliminary, but demonstrate the possibility of combining natural language, agentic workflows, and domain-specific structural data structures as a first step toward AI-assisted structural co-design.
agentai agentagentic - arxiv:2610.31111 · cs.CVWhere Do Long Captions Fail? Position-Aware Diagnosis and Reinforcement Learning for Detailed Image CaptioningYuanze Hu, Zhichao Yang, Junwei Jing, Xin Yang +8
Detailed image captioning is usually framed as an amount problem: a model should mention more visual facts while avoiding hallucination. However, autoregressive VLMs generate captions as ordered sequences, and later generations may be produced under weaker or noisier visual grounding. We therefore argue that long-form captioning should be studied not only by what facts are mentioned, but also by where they appear. We introduce a position-aware rubric representation that decomposes dense human references into atomic visual claims, verifies generated captions against those claims, and induces a quartile-wise composition of supported, hallucinated, and generic content. This lens reveals two qualitatively different failures in long captions: models tend to produce unsupported specificity in the middle, while the final portion often escapes into vague, non-evidential language rather than simply accumulating more hallucination. A same-reference human control does not exhibit these patterns; additional human audits examine the visual validity of the induced labels. Motivated by this diagnosis, we propose PAGER (Position-Aware Grounded Evidence Reinforcement), a GRPO training objective with separate signals for supported fine-grained middle evidence, evidence-bearing tail continuation, claim coverage, and content-aware length control. The result is a closed loop: the same ordered claim representation identifies where long captions fail and supplies the position-specific reward needed to repair those failures. Experiments on detailed-caption benchmarks show that PAGER improves position-wise grounding and pairwise caption preference while preserving competitive global caption quality.
benchmark - arxiv:2610.31107 · cs.CLLearning What to Ask for Tomorrow: Active Preference Acquisition for Long-Term Personalized Web AgentsYiqi Wang, Zizhi Zeng, Jiaqi Zhang, Yiqun Duan +2
Web agents increasingly serve the same user across repeated online tasks such as booking travel, shopping, and scheduling, so their long-term usefulness depends on how well they know the user's preferences. Passive memory and personalization methods cannot obtain preferences the user never reveals, and clarification methods that value a question by its benefit to the current task overlook the benefit of reusing the answer in later tasks. We study Active Preference Acquisition (APA): which missing preferences are worth the cost of asking? We define the Prospective Personalization Value (PPV) of a missing preference as its expected cumulative utility gain over future tasks minus the effort and disclosure-risk cost of asking, and select questions under a per-interaction budget. We characterize when cumulative reuse justifies acquisition despite insufficient single-task benefit, and evaluate this regime with an initial-profile diagnostic. APA computes the counterfactual gain of each candidate preference on public task prototypes with the agent's own task solver and aggregates it by the expected reuse of each task family. We evaluate it on ActiveWebPref, a controlled longitudinal benchmark with hidden but queryable user profiles and exact task utilities. At a horizon of 20 tasks, 34.4% of budget-feasible valuable missing preferences are amortization-dependent; APA acquires 80.3% of them, whereas current-task value of information acquires 5.1%. Under per-user asking-cost caps matched to APA's realized expenditure, APA achieves higher net utility than four heuristic selection baselines. In our cost setting, asking about every unknown preference yields lower net utility than not asking, while APA recovers 84.2% of its utility gain with 27.6% of its questions. The value of acquiring a user preference should account for its reuse across future tasks, not only its benefit to the task at hand.
memorybenchmark - arxiv:2610.31105 · cs.AIWho Is the Influencer? Consensus Dynamics in Multi-Agent SystemsGouki Minegishi, Jingyuan Feng, Hiroki Furuta, Takeshi Kojima +2
Multi-agent LLM systems are increasingly used in practice, but their behavior is complex and hard to understand. We design a simple task to ask which model has the most influence on the final consensus when several models are mixed in one system. Experimentally, across seven model families, we find that the opinions of the other models are absorbed into the opinion of the stronger model, the one with more parameters. This happens even though the task requires no pretraining knowledge and has no correct answer. To understand this phenomenon, we interpret the agents as weighted particles moving on a simplex. Larger models carry a larger weight and therefore occupy a deeper well in the energy landscape. A large weight means strong in-context learning: a model with a large weight amplifies the majority of its context. Larger models are generally better at in-context learning, and this is what makes the larger model the influencer. We hope that this work helps to interpret multi-agent systems, whose behavior is becoming too complex for humans to understand.
multi-agentagent system - arxiv:2610.31098 · cs.CLRIFT: KV Cache Compression for Draft Models with Complementary Compressibility TrainingDor Tsur, Maor Ashkenazi, Maximilian Golub, Peter Hicks +6
Speculative decoding is a popular decode acceleration method. As such, it is particularly relevant to long-output generation, which requires an increasingly large KV cache. KV cache compression offers a way to constrain costs in long-context drafting. However, compression methods that focus on prefill information can lose relevance as the output length grows. We term this phenomenon decode staleness and show that it degrades draft model accuracy under selection-based KV cache compression. We introduce RIFT, a family of selection-based KV cache compression strategies that balance prefill information and recent generated tokens. We show that refreshing the KV cache selection during token generation significantly improves the drafter's quality at the cost of additional decode latency. To alleviate the refresh costs, we propose an adaptive KV refresh strategy that reduces the number of score calculations while retaining the drafter's quality. We further demonstrate how a compressibility-driven drafter training phase, based on sparse attention masks, can improve the drafter's quality under a compressed KV cache. Combined with compressibility training, RIFT enables Qwen3.5-4B, drafting for Qwen3.5-122B, to retain over 98% of its full-cache acceptance length on 32K-token inputs, 4K-token outputs, and an 8K-token KV cache budget.
long-context - arxiv:2610.31097 · cs.AICo-Recursive Semantic Self-Extension: A Mathematical Foundation for Open-Ended IntelligenceSimon Hegelich, Bardh Prenkaj, Habiba Sarhan, Morteza Shahrezaye +3
Artificial intelligence has advanced through scale, data, and computation, yet remains bounded by its representational framework. We ask what makes intelligence open-ended. Every inferential system selects meaningful distinctions, limiting what it can learn. Within a fixed purpose, performance reaches a ceiling set by its semantic domain. Teleological succession, in which purposes arise from earlier ones, faces a similar limit: Boolean-closed purpose-transformations exhaust their reachable semantic territory. We shift from purposes to inferential systems. Drawing on Kant and Gödel, we model cognitive states as partial evaluations over a formal language. Unresolved propositions may be undecided or undecidable. Reasoning can resolve the former but not the latter, leaving a frontier for extension. Non-trivial reductions of undecidable propositions expose undecidable intermediate premises whose reflective reification supports multiple decisive extensions. Finite trace-guided generation remains fallible, requiring meta-level selection and polycontextural exploration. Using Kolmogorov complexity, we show that no architecture with a fixed state description, transition mechanism, and initial condition can generate histories incorporating irreducible algorithmic information. Because physical systems are fixed in this sense, we examine collectives. Coupling finitely many finite agents does not remove the bound. Extension requires persistent external memory preserving semantic content and historical coordinates. Collective generative capacity depends on balancing overlap for transfer with difference for generation. We propose semantic self-extension as a design principle: intelligence should not merely resolve ambiguity within existing meanings, but transform unresolved structures into new spaces of meaning, generating new distinctions, questions, representations, and purposes.
memoryexternal memory - arxiv:2610.31087 · cs.CVI$^2$-KAN: Interpretable Intra- and Inter-Modal Multi-Pathway Representation Learning for Multimodal Sentiment AnalysisXiaoting Xie, Feng Qiao, Xianwei Zhang, Hailong Chen +4
Multimodal sentiment analysis (MSA) aims to infer sentiment by integrating linguistic, acoustic, and visual information. However, many fusion methods integrate modality-specific representations and cross-modal interactions into joint representations, making it difficult to determine whether predictive utility arises from a modality itself or from its interactions with others. More fundamentally, a modality's standalone predictive utility may not faithfully reflect its interaction-conditioned predictive utility. To address this issue, we propose I$^2$-KAN, an interpretable multi-pathway framework that explicitly maintains intra-modal and pairwise inter-modal information in separate structural pathways. Each pathway is refined by a Task-Decoupled Linear-Gated Denoising Bottleneck (TDL-GDB) to learn compact and task-relevant representations, followed by structured Kolmogorov-Arnold Network fusion that retains explicit pathway associations and supports multi-level interpretation. Extensive experiments on CMU-MOSI, CMU-MOSEI, and CH-SIMS v2 demonstrate strong overall performance. Ablation and interpretability analyses consistently show across the three benchmarks that low standalone predictive utility does not necessarily imply low interaction-conditioned predictive utility: a modality that appears weak in isolation can remain informative through pairwise interactions. These findings highlight the interaction-dependent nature of multimodal utility and suggest that modality utility should be characterized from both standalone and interaction-conditioned perspectives.
action-conditionedbenchmark - arxiv:2610.31083 · cs.AIMQAScore: Fine-Grained Text-Music Alignment Evaluation via Large Audio Language ModelWei-Jaw Lee, Chun-Ping Wang, Xuanjun Chen, Yi-Hsuan Yang
Text-to-music alignment is commonly evaluated by the global similarity between text and audio embeddings, which collapses all described attributes into a single score and obscures which ones are missing. We instead introduce MQAScore, which verifies each atomic musical concept along five clip-level dimensions (genre, instrument, mood/theme, vocal, and usage) by the probability that a large audio language model (LALM) answers affirmatively to a binary question. To assess evaluator reliability, we construct verification labels from presence/absence annotations and LLM-guided concept substitutions, and propose the Sensitivity-Invariance Ratio, which measures how selectively an evaluator responds to meaning-changing substitutions. Experiments show that concept-level scoring improves both question-answering and similarity-based evaluators, and MQAScore can be a more reliable and interpretable fine-grained complement to global similarity for evaluating text-to-music alignment.
evaluator - arxiv:2610.31080 · cs.LGnnInteractive Agent: General-Purpose AI Agents as 3D Image AnnotatorsFabian Isensee, Constantin Ulrich Harsy, Maximilian Rokuss, Benjamin Hamm +2
Interactive 3D segmentation models such as nnInteractive turn a few spatial prompts (points, scribbles, boxes) into accurate volumetric masks, but a human still does the annotation work: finding each structure, placing prompts, checking the result and correcting it. General-purpose AI agents can now interpret images, call tools and write code. We show that such agents can take over this work without additional training: the user makes a request in plain language, and the agent works out what has to be segmented, locates the targets, prompts nnInteractive, inspects the 3D result, corrects it and responds to feedback. Asked for 'the muscles used for chewing', an agent identifies and segments the masseters and pterygoids; asked to count liver lesions, it writes its own code to find candidates and splits a merged lesion on request. Our open-source nninteractive-agent provides nnInteractive as tools over the Model Context Protocol (MCP), together with a skill that encodes how experienced users prompt it. On a benchmark spanning CT, MRI, PET and micro-CT of humans, a mouse and an ant (14 images; 51 targets from organs and tumours to muscles, foot bones, vessel trees and an insect brain), Claude Code and Codex reach a mean Dice of 0.81 and 0.84 from the request alone; when told where each target is, both reach 0.85 and prompt nearly as effectively as an oracle that places prompts using the reference segmentation. A text-prompted 3D model (VoxTell) is more accurate on most standard radiological targets but less robust: it returns an empty mask for one of them and fails on the leg muscles, the foot bones, the mouse and the ant, where the agents remain accurate. Code: https://github.com/MIC-DKFZ/nninteractive-agent
agentai agentbenchmark - arxiv:2610.31075 · cs.LGWhen Forgotten Facts Return: Relearning-Aware Unlearning of Knowledge Graph Embeddings via Tangent KernelsParnian Fazel, Mohammad M Maheri, Hamed Haddadi
Approximate unlearning answers a deletion request on a knowledge-graph embedding by editing entity embeddings instead of retraining from scratch. These embeddings keep being updated as new facts arrive. Under this ordinary continued training, forgotten facts recover more than in the exact-unlearning baseline, a model retrained without them and given the same continued training. We trace this relearning to entity embeddings shared by forgotten and new facts. To our knowledge, this is the first study of relearning in knowledge-graph embeddings. Relearning-robust methods developed for deep networks do not account for the graph structure underlying this recovery. We propose Kernel-Anticipated Unlearning (KAU), which uses the tangent kernel to model how continued training changes forgotten facts' scores through these shared parameters. KAU expresses this first-order response in closed form and differentiates through it inside the forgetting loss. The objective steers edits toward directions that continued training reverses less. We compare KAU with state-of-the-art knowledge-graph unlearners and relearning-robust methods under identical continued training, measuring leakage beyond the exact-unlearning baseline. Across three backbones, KAU reduces the strongest knowledge-graph unlearner's leakage by 46-63% on FB15k-237 and 50-82% on CoDEx-L, at higher retained utility.
knowledge graph - arxiv:2610.31066 · cs.AIDual Exam: Controlled Evaluation of Multimodal Agents and World ModelsZhefan Rao, Meng Chu, Xuan Zhang, Jingfan Chen +79
World Models (WMs) can support Multimodal Agents (MAs) through simulated experience, while Agent trajectories can inform World Model refinement. Evaluating this interaction requires separate evidence about action selection and state evolution. We introduce Dual Exam, which tests Agents under fixed environment response mechanisms and World Models under fixed action interventions across physical, digital, social, and scientific domains. The Agent track compares 10 configurations on 1,088 items in 15 reporting categories. The World Model track contains 400 cases across eight scenarios; 15 configurations have full coverage and one covers six scenarios. GPT 6 Astra leads the case-weighted Agent Index at 31.3, and GPT-Image-2 leads the full-coverage WM Overall at 21.7. These scores use different tasks and metrics and are not directly comparable. A 150-case pilot of one coupled pairing records zero task successes and motivates component-level diagnosis. Results reveal uneven task performance and failures to preserve task-relevant state. Their value for predicting coupled success or sustained reciprocal learning remains to be established.
world modelagent - arxiv:2610.31064 · cs.ROAstroBench: Benchmarking Free-Floating Capture and Space Robotic ManipulationZizhe Wang, Jia Deng, Xin Lu, Zehao Jia +2
Capturing a loose object inside a spacecraft requires a robot to approach a target that can translate and rotate, and whose motion changes when contact is made. These conditions motivate evaluation beyond manipulation of supported objects. We introduce AstroBench, an open benchmark for free-floating capture and space robotic manipulation with a fixed-base arm. The task suite combines four initial-motion tracks with sample-storage operations based on MELFI hardware and cap and captive-screw servicing based on the Robotic Refueling Mission. The two workcells provide nine task definitions and 29 task-target combinations. A metadata-filtered Objaverse inventory supplies 8,414 object instances for asset expansion, alongside five cabin visual contexts and versioned task instructions. We report two acquisition-focused experiments using existing checkpoints. A frozen $π_{0.5}$ DROID joint-position policy captures 2 of 200 static targets and 4 of 200 moving targets across twenty object identities. A cube-specific ACT policy captures 8 of 10 static targets and 6 of 10 moving targets in a separate experiment. Each experiment retains its recorded observation, action and capture configuration. The results document difficult direct transfer for the tested generalist checkpoint and successful task-specific capture behavior, while identifying concrete settings for studying motion-aware perception and contact-sensitive control. Code, task definitions and editable workcell assets are available at https://github.com/AstroUpLab/AstroBench.
manipulationbenchmark - arxiv:2610.31062 · cs.AISCOPE: Compiling Natural-Language Policies into Neurosymbolic Predicate ProgramsArnon Mazza, Elad Levi
Policy enforcement for LLM agents has shifted from single-turn content moderation to compound, application-specific policies evaluated over multi-turn transcripts and tool-call trajectories. Fine-tuned guardrail models are efficient but degrade on compound policies, and while LLM-as-a-judge approaches achieve higher accuracy, they incur high inference cost and tend to produce unstable predictions. Neurosymbolic methods offer a natural alternative, decomposing the policy into narrow neural decisions composed under explicit logic. However, existing symbolic layers lack the required expressiveness and are optimized against fixed data rather than the policy itself. To address this, we introduce SCOPE, which compiles a policy into an executable program of neural predicates executed by small models, deterministic extractors, and code that composes their outputs. The program is optimized jointly with its training data, using adversarial generation of counterexamples that target where the program deviates from the policy. We evaluate SCOPE on a benchmark of real-world policy tasks against prompt optimization, task decomposition, and workflow generation methods, achieving the highest average accuracy across tasks and models. Programs optimized with the weakest executor transfer to other models without further optimization. This allows them to be deployed in production on inexpensive models, where they reach accuracy comparable to frontier reasoning models. Unlike these monolithic models, SCOPE produces substantially more stable predictions, and each verdict is interpretable through the predicate outputs that produced it. Additionally, we validate the key components of our method through an ablation study. SCOPE thus enables reliable, interpretable, and efficient policy enforcement at production scale.
llm agentbenchmark - arxiv:2610.31060 · cs.AIDiagnosing the Explicit-to-Applied Theory of Mind Gap in LLMsPrajvi Saxena, Sabine Janzen, Wolfgang Maass
As large language models are deployed in high-stakes roles such as therapy, education, and personal assistance, their social cognition must be assessed reliably. Current Theory of Mind (ToM) benchmarks report aggregate scores that conflate two distinct competencies: explicit ToM (recognizing what someone believes) and applied ToM (using that belief to predict behavior). Building on this distinction, we investigate the gap and its recoverability using four structurally diverse benchmarks. Aggregate reporting obscures uneven competence: on SimpleToM, Llama-3.1-8B scores 51.5 percent overall while recognizing mental states at 99.4 percent and predicting behavior at only 28.9 percent. To assess recovery, we compare structured prompting, matched in-context learning, and our proposed Belief-State Test-Time Training (BSTTT), which adapts lightweight parameters on answered support questions through action reconstruction. Across Qwen2.5-7B, Llama-3.1-8B, and GPT-oss-20B, recovery depends on the model, task, and intervention. On SimpleToM, reminders combined with BSTTT improve Llama's behavior prediction by 40.8 percentage points; reminders account for most of the overall improvement, with BSTTT adding 1.3 to 5.5 points beyond reminders across models. Contextual examples often outperform weight updates, while BSTTT remains stronger in specific model-benchmark settings. Temporal belief tracking remains difficult despite partial recovery. These findings show that aggregate ToM scores can overstate applied competence while hiding performance recoverable through targeted assistance. We recommend decomposed reporting of explicit and applied accuracy alongside intervention comparisons that establish which failures are recoverable and under what conditions.
benchmark - arxiv:2610.31054 · eess.SYCross-Region Cooperation of UAV Swarms Towards Air-Ground ISAC via Hierarchical Multi-Agent Reinforcement LearningLinghui Miao, Nhan Thanh Nguyen, Shijian Gao
The limited communication resources and sensing perspectives of a single uncrewed aerial vehicle (UAV) motivate UAV swarms for integrated sensing and communication (ISAC) in air-ground networks. However, in multi-region scenarios, regional UAV allocations may not match vehicle demand, while local observations may lack sufficient diversity for accurate sensing. To address these issues, this paper proposes a cross-region cooperation framework for UAV swarms. For communication, cross-region UAV reallocation reduces service disparities by matching resources to vehicle demand. For sensing, combining UAV observations from different regions increases observation diversity, while adaptive handshaking guided by marginal Cramér--Rao bound (CRB) utilities balances phase synchronization gains against handshaking energy consumption. Joint optimization of UAV positions and transmit power within each region then further improves communication service and sensing performance. These coupled decisions are coordinated by a hierarchical multi-agent reinforcement learning (H-MARL) approach through slow-timescale inter-region UAV reallocation and fast-timescale intra-region UAV control. Under the considered simulation settings, reallocation can improve quality of service (QoS) by over 20 percentage points compared with retaining the initial allocation under severe resource-demand mismatches. Cooperative sensing with adaptive handshaking incurs only about a 5.0% increase in mean CRB relative to perfect synchronization.
multi-agent - arxiv:2610.31047 · cs.AIHarm Is Behavioural, Moderation Is Lexical: How Language and Culture Decide Whom Automated Safety Systems Protect and Whom They SilenceGabriel Lars Sabadin, Dr Nicola Harding, Sarah Kingston
Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked. Automated systems decide at scale which messages are harmful, and they decide unevenly. Classifiers label African American English as offensive up to twice as often as other English, the accounts of prominent drag queens can score as more toxic than those of white nationalists, and safety behaviour that holds in English collapses in low-resource languages. We argue that over-flagging and under-protection share one cause: deciding harm from vocabulary and surface form rather than from what a message does. We organise the evidence into seven failure modes and argue, as testable hypotheses, that harmful behaviours (a request for secrecy, a demand for payment, an attempt to isolate someone) travel across languages better than the words that express them. We propose an evaluation protocol that makes the gap between the best- and worst-served language communities a headline result, and show that the naive gap is inflated under equal service and that its usual bootstrap interval is unreliable. Applied to five open systems in eleven languages, from a keyword lexicon to two current multilingual guard models, the protocol finds that none reaches native-speaker accuracy at separating harmful function from harmless form (51 of 55 system-language pairs fall short); that greater capability raises discrimination in nine languages while leaving Arabic and romanised Hindi near chance, so the gap between the best- and worst-served languages grows with capability; that one system both over-flags and under-protects the same language most; and that a language a system reads worst can look the most fairly treated when only false positives are reported.
evaluation protocol - arxiv:2610.31046 · cs.AIConflict-Aware Logit Adapters for Utility-Preserving Anti-DistillationQuoc Phong Dao, Binh Anh Nguyen, Linh Ngo Van, Thanh-Toan Do +2
While reasoning traces improve model capability, they also provide supervision signals for downstream distillation. Initial efforts to mitigate this issue rely on decoding-time interventions, but they incur high inference overhead and can harm the teacher model's natural fluency. We propose Conflict-Aware Logit Adapters (CALA), a lightweight training-time defense. CALA attaches a small residual adapter to a frozen teacher model and trains only the adapter. A proxy student identifies positions where imitation is most sensitive, and the adapter perturbs the output distribution at these positions while remaining within high-probability regions. A conflict-aware optimization strategy balances anti-distillation and utility objectives, while regularization preserves the teacher's performance and fluency. At inference time, CALA incurs modest overhead and requires no auxiliary model. Experiments on reasoning benchmarks show that CALA substantially reduces successful student imitation while maintaining near-original teacher performance and generation quality. These results establish CALA as a practical and efficient alternative to decoding-time approaches.
benchmark - arxiv:2610.31041 · cs.AIAlgorithmic drift in generative adaptive learningFiruz Kamalov, Hana Sulieman, Ahmed El Sayed, Sivaraj R. +1
Generative artificial intelligence promises to democratize personalized tutoring, yet the longitudinal stability of adaptive systems remains under-theorized. Current ethical discourse focuses on static bias. This paper argues that a distinct threat lies in algorithmic drift: a feedback loop in which minor, demographically asymmetric observation errors compound over time into systemic inequities. We make three contributions. First, we formalize algorithmic drift as a coupled learner--tutor dynamical system. Second, we corroborate this theory with an agent-based Monte Carlo simulation in which either a transparent online tracker or an LSTM deep-knowledge-tracing network drives an IRT-based curriculum policy. Third, we introduce the temporal fairness stabilizer (TFS), a targeted intervention that periodically re-probes learner capability, and benchmark it against standard exploration baselines with explicit group-fairness metrics and replication-level inference. Within the modeled environment, TFS closes the achievement gap where generic exploration does not, and we characterize the conditions under which it succeeds. We separate what is established analytically, what is demonstrated in simulation, and what remains to be validated with real learners. Our results reframe algorithmic fairness in education as a temporal property of a closed loop rather than a deployment-time checkbox.
benchmark - arxiv:2610.31034 · cs.AICANNBot-CAKE: Collaborative Agentic Kernel Evolution for Ascend NPUDeng Pan, Xue-Jian Gao, Junyi Fan, Bin Du +7
Efficient custom operators are essential for modern deep learning workloads, yet developing high-performance kernels for Ascend NPUs remains slow, expert-intensive, and difficult to automate. Although AI agents now approach expert performance on GPU kernel benchmarks, current models remain less capable of generating AscendC kernels for NPUs, potentially reflecting the relative scarcity of relevant training data. We thus present CAKE, a CANN-native system for deployable custom operators that combines adaptive multi-DSL routing, separate Generation and Evolution modes, and reviewed knowledge reuse through CAKE Book and a Self-Evolving Agent. On 53 CANN-Bench v1.1 operators on Ascend 950PR, CAKE's aggregated results achieve $100\%$ case pass rate on the standard set (1,060/1,060) and $98.28\%$ on the hidden set (4,167/4,240), with an overall mean score of 78.63 out of 100 and an average performance speedup of $1.334\times$ over the baseline. Evolution then pushes correct operators further: EngramGateFusion cuts aggregate device time by $46.6\%$ over 20 rounds and roughly halves the mean roofline gap. These results offer a practical path toward closing Ascend's AI-for-kernel capability gap.
ai agentagenticself-evolvingbenchmark - arxiv:2610.31024 · cs.LGNN-Based Transistor Modeling for Ultra-Fast SPICE Simulation and Standard Cell Library CharacterizationRodion Novkin, Hussam Amrouch
Neural network (NN)-based transistor models have proven to be a promising solution to accelerate device modeling. Although these models demonstrated remarkable speedup in circuit simulations, they were not rigorously tested in complex tasks within standard EDA tool flows for chip design, which require very high transistor model precision to provide accurate results and converge properly. To investigate their full potential and promise, we develop NN-based transistor models that accurately predict the drain current and various charges for a 3nm nanosheet, and employ the models for a standard cell library characterization. This enables us to evaluate the accuracy and speedup under a very wide range of SPICE simulations, benchmarking our models against an industry-standard implementation. Since standard cell library characterization involves thousands of SPICE simulations, it serves as a great benchmark for transistor model quality and consistency. To further evaluate model accuracy, we also perform analog- and digital-level simulations for various complex circuits. Our experiments reveal that NN-based models are a valid alternative to existing transistor compact models, capable of producing accurate delay and power estimates with a much higher speed and convergence time compared to the conventional Verilog-A-based industry-standard (BSIM-CMG) transistor compact model. Our NNs achieve sub-0.11% and 0.76% errors for delay and total power evaluation on a wide range of circuits, while taking up to 8 times less time during the standard cell library characterization. Additionally, we provide comprehensive information and guidance on NN accuracy, unveiling the relationship between NN size and its effect on circuit simulation precision.
benchmark - arxiv:2610.31020 · cs.RODemonstration-Based Robotic Feeding Using Dynamic Movement PrimitivesKenzhi Iskandar Wong, Kitti Thamrongaphichartkul, Sri Harsha Turlapati, Yanshuo Lu +9
This paper presents a demonstration-based robotic feeding framework that integrates human demonstration, Dynamic Movement Primitive-based motion planning, vision-based scene perception, and robot control. A human operator first demonstrates the scooping and feeding motions while the position and Euler-angle orientation of the spoon are recorded using a motion-capture system. The recorded trajectory is segmented into scooping and feeding motions, which are encoded using separate Dynamic Movement Primitives. Before task execution, an RGB-D camera estimates the participant's mouth position, while two AprilTags are used for robot-frame registration and tray localization. The detected tray pose is used to position the scooping motion, whereas the detected mouth position defines the positional goal of the feeding motion. The approach, scooping, and feeding trajectories are then executed sequentially by a Kinova Gen3 manipulator through a ROS~2-based framework. A real-world experiment demonstrates the feasibility of the integrated pipeline and shows that the learned spoon motions can be reproduced within the detected experimental scene.
manipulator - arxiv:2610.31019 · cs.CLChunk Recomputation under Real Document Revisions: Measuring Update Cost and Retrieval Quality of Fixed, Recursive, and Content-Defined ChunkingYuning Bian, Lele Xu, Yifan Yuan, Xinyu Yin +1
Chunking strategies for retrieval-augmented generation are almost always evaluated on a static snapshot, yet the documents underlying a knowledge base continue to change. Because chunks are embedded independently, a boundary that moves after an edit forces unchanged text to be re-embedded. We measure this update cost on 120 pairs of adjacent Chinese and English Wikipedia revisions for three stateless chunkers (non-overlapping fixed-size, recursive paragraph-first, and content-defined chunking) under a strict byte-identical reuse rule and a frozen dense encoder. On the 116-article common-success pool, with frozen, length-calibrated parameters, fixed-size chunking recomputes 57-68% of its full new-version embedding-input workload, recursive chunking 38-56%, and content-defined chunking 43-54% (it failed on 4 of 62 Chinese articles), while at these parameters mean evidence recall under a fixed 2048-token retrieval budget differs by less than two percentage points within either language. A cost-only parameter grid shows that the ranking of the two content-aware chunkers changes with chunk-length settings, whereas non-overlapping fixed-size chunking has the highest recomputation cost in every length-matched condition. Controlled edits and a verbatim-presence diagnostic support boundary drift as one contributor to recomputation of text already present in the previous version. Update cost is therefore a substantial, parameter-sensitive dimension that static retrieval evaluations do not expose and that should be reported alongside retrieval quality.
retrieval-augmented - arxiv:2610.31012 · cs.AIEvaluating Artificial Wisdom (AW): A Trajectory-Based Framework for Context-Sensitive Decision RegulationMd. Jalil Piran
Capability benchmarks measure what large language models can do, but not whether their recommendations remain appropriate as decision conditions change. This paper proposes a trajectory-based measurement framework and prospective validation protocol for artificial wisdom, defined behaviorally as context-sensitive regulation of judgment. Versioned reference specifications determine admissible responses and transition materiality before evaluation. Seven stage-level criteria assess judgment under current conditions, while two transition components assess material-update adequacy and commitment-preserving stability. Stability requires prior adequacy; switching within registered commitment classes is permitted, and persistent error earns no stability credit. A nine-component profile is primary, with a secondary bottleneck score and separate critical-failure reporting. Three simplified simulations examine precision, bottleneck estimation, and sensitivity to model-family count. A single-criterion simulation met its precision criteria at 24 scenarios per domain. Near a bottleneck tie, the estimated minimum had a downward bias of 0.55 points at that size. A simplified incremental-prediction simulation detected a Brier improvement of 0.027 with probabilities of .508 and .742 using eight and twelve families, respectively; it does not establish power for the operational protocol. An exploratory predecessor-rubric study comprising 71 evaluation records illustrates aggregation sensitivity but does not validate the proposed instrument. The twelve-family target remains provisional pending feasibility assessment and calibration of the complete analysis. Whether the profile predicts independent decision outcomes beyond capability remains an open empirical question.
benchmark - arxiv:2610.31010 · cs.RORevisiting Distribution Matching for State-to-Vision Policy Distillation in Laparoscopic Robotic ManipulationMingyu Sheng, Jianan Fan, Dongnan Liu, Guoyan Zheng +2
End-to-end control policies trained via Reinforcement Learning (RL) have emerged as an alternative to conventional surgical robotic control approaches, with recent studies exploring state-based policies trained on privileged state observations. However, such approaches rely on precise physical states that are difficult to obtain reliably in surgical environments. Vision-based RL provides an alternative by learning manipulation policies directly from RGB observations, avoiding explicit state estimation and manually engineered perception pipelines while enabling policies to acquire task-relevant visual representations. Nevertheless, obtaining vision-based policies via pure RL methods for surgical manipulation remains challenging. Thus, imitation learning (IL) becomes another solution, yet it requires a large number of expert demonstrations, which are difficult to obtain in minimally invasive surgery. In this work, we propose a framework for state-to-vision policy distillation in laparoscopic robotic manipulation. Specifically, a state-based policy with privileged observations is employed as the teacher, and a vision-based policy learns from RGB observations and robot proprioceptive information, which requires less expert demonstrations than IL and introduces superior performance than RL methods. We conduct extensive experiments on various surgical manipulation tasks in the simulated environment. Experimental results show that, in the context of laparoscopic robotic manipulation, training vision-based policies with general-purpose RL methods is challenging, whereas our proposed state-to-vision policy distillation leads to more consistent and stable performance across tasks.
manipulation - arxiv:2610.31005 · cs.AIMitigating Saliency Collapse: Robust Saliency-Aware Long-Text Image-Text AlignmentQiuyu Kong, Zanxi Ruan, Marco Cristani, Yiming Wang
Long-text image-text alignment is essential for fine-grained scene understanding, yet existing methods mainly focus on extending context length or improving local alignment. We identify an overlooked failure mode, saliency collapse, where models over-rely on text that features visual prominence while underutilizing non-salient context. Consequently, retrieval performance degrades sharply when salient content is reduced, despite that the non-salient context remains informative. To address this, we propose RoSA, a saliency-aware finetuning framework that decomposes image-text pairs into complementary salient and contextual views, enabling joint learning of prominent and contextual semantics. We propose a modality-independent decomposition strategy to combine object-level visual saliency estimation with LLM-based text decomposition to bridge vision and language. We employ auxiliary objectives to explicitly align these views, ensuring robust representation with both salient and contextual features. RoSA greatly improves long-text retrieval and demonstrates strong robustness to saliency collapse across benchmarks. Moreover, our auxiliary objectives are plug-and-play, bringing consistent gains to existing tuning methods. Code and checkpoints are available at: https://github.com/intelligolabs/RoSA
benchmark - arxiv:2610.31003 · cs.AIAgri-SCHOLAR: Simulation-Grounded In-Season Rolling Advisories for Agricultural Crops Using LLM-Bounded OptimizationGeetha Charan, Vedant Balasubramaniam, Manojkumar Patil, V. Priyanka +4
Existing agricultural advisories to farmers suffer from several limitations: (a) remain largely expert-driven, (b) do not adequately account for in-season weather, and (c) do not scale to individual farms. In this paper, our objective is to close these gaps by developing Agri-SCHOLAR (Simulation-Grounded Crop Harvest Optimization with LLM-Guided In-Season Advisory Rollouts). Agri-SCHOLAR is a rolling, in-season advisory framework that issues, at intervals of every few days or every week, a personalized advisory using the following information: soil and crop properties, observed weather to date, short-range forecast of weather, and a season extrapolation of weather based on historical analogs. The novelty and depth of Agri-SCHOLAR lies in each advisory being produced by a closed-loop system which couples multiple LLM agents with a powerful crop simulator (such as APSIM in our case). The simulator crucially serves as an evaluation oracle. In Agri-SCHOLAR, the LLM agents propose agronomically sound and feasible bounds on actionable parameters and an optimizer searches within those bounds against a multi-objective reward spanning yield, resource utilization, and economic feasibility. We conduct a detailed 11-year analysis of the maize crop in two selected locations of South Asia and show that Agri-SCHOLAR consistently outperforms the region's standard packages of practice and satisfactorily addresses the limitations of existing agricultural advisories.
llm agent - arxiv:2610.31000 · cs.AISecLog-Bench: A Scalable Benchmark for Cybersecurity Log Reasoning with Scaffolded EvidenceYifan Wang, Dongliang Xu, Xiaoyong Li, Jing Li +2
Large language models (LLMs) have shown strong potential for cybersecurity, leading to growing interest in LLM-based automated cybersecurity log analysis. However, current datasets (1) are limited in scale; (2) rely mainly on single-source evidence; and (3) have only binary or multi-class attack labels, without dense annotations of the reasoning process. To address these limitations, SecLog-Bench has three key features. First, we build an extensible simulation platform that covers 29 attack scenarios, supports repeated execution and the integration of new scenarios, and enables scalable generation of cybersecurity logs with attack-stage labels. Second, we use multi-source logs, including audit logs, application logs, and network traffic, to jointly analyze cyber attacks. Third, we annotate each sample with the essential context for cybersecurity log reasoning and a structured step-by-step evidence chain. We evaluate eight mainstream LLMs on this benchmark for post-detection attack-progress reasoning. Results show that, when provided with scaffolded evidence, current LLMs exhibit modest capability in attack-progress reasoning, with the best model achieving only 44.44% accuracy and 45.07% Macro-F1. Their reasoning remains unreliable; i.e., the highest Stage Reasoning F1 (SRF1) score is only 54.87%. Moreover, even correct predictions may be accompanied by flawed reasoning, and reasoning quality is often misaligned with prediction accuracy.
benchmark - arxiv:2610.30991 · cs.ROUniFlowWAM: A Unified World-Action Model with Visual Semantics and 4D GeometryJialin Tian, Zhuofan Li, Xinkai Wang, Bokai Lin +4
Robot demonstrations record what the robot saw and what it did. They do not specify how the 3D world changed in between. Scene-level 3D flow describes it by tracking 3D scene points, providing the world representation with 4D geometry. Prior work has used flow to improve robot policies. Flow served as an intermediate prediction that a separate model converts into actions, as a prior injected at inference time, or as an extra output learned together with the action. In all of these designs, flow stays outside the policy, as a separate module or as an extra prediction head. We introduce UniFlowWAM, a Unified World-Action Model that makes 3D scene flow a third modality on par with vision and action. Its world representation therefore covers both visual semantics, through future visual features, and 4D geometry, through 3D flow, and flow receives its own supervision. UniFlowWAM is a diffusion transformer with one expert per modality, and the three experts exchange information through joint attention. We train UniFlowWAM under a ten-objective role grid, where each objective assigns every modality exactly one of three roles, namely target, condition, or absent. The same model thus serves as a policy, as forward and inverse dynamics, and as a predictor of future visual features or 3D flow. We evaluate UniFlowWAM on RoboCasa-GR1, LIBERO, and LIBERO-Plus in simulation, and on four real-world bimanual manipulation tasks. On RoboCasa-GR1, UniFlowWAM raises the success rate of its base model without flow from 52.1% to 56.4%. It preserves in-distribution success on LIBERO at 98.1% and reaches 87.7% zero-shot on LIBERO-Plus. On the real robot, it achieves 87.0% average success, compared with 68.5% for the strongest baseline.
manipulationlibero - arxiv:2610.30989 · cs.AIColaBench: Evaluating Memory-Guided Knowledge Acquisition in Personal AgentsWenjie Zhang, Zhichao Shi, Xuhui Jiang, Xiaojun Wu +3
Personal agents need to draw on long-term interaction history to retrieve and apply external knowledge suited to a user's current request. Even an answer grounded in relevant documents can be unsuitable when it overlooks the user's circumstances. This creates two connected challenges: using historical conditions to identify the knowledge needed for a new request, and applying that knowledge to the user's situation. Existing benchmarks advance memory recall, personalized retrieval, and memory-guided action, while the joint evaluation of these two challenges remains underexplored. We introduce ColaBench, a synthetic benchmark in health and daily living, with 349 tasks across 10 Profiles and a knowledge base of 13,033 documents. Its Evidence Graph connects personal conditions to their disclosures across public interaction sessions, records prior document exposure, and links answer requirements to personal and newly acquired document evidence. We also introduce MemKB, which uses state summaries to inform queries and rule application, and memory-document links to retrieve knowledge associated with the user's history. MemKB achieves 86.25% Accuracy with an average of 4.07 interaction rounds per task. Task-level analysis finds larger gains on tasks integrating information across multiple historical Facts; error analysis further identifies unsupported personal assumptions and rule misapplication as obstacles to using retrieved evidence. ColaBench provides traceable tasks and an evaluation framework for studying how memory guides knowledge acquisition and rule application.
memorybenchmarkevaluation framework - arxiv:2610.30988 · cs.ROIn-Context Learning World Models for Robot Dual-Arm Manipulation of Soft Garment with Partial ObservationsHengyi Sim, Quentin Rouxel, Dennis Hong, Darwin Caldwell +1
Robust manipulation of soft objects, such as garments, is a critical skill for service and household robots to adopt, but remains challenging due to its high-dimensional state space, and material-dependent deformations. Humans address this challenge by using mental world models that "imagine" the shape and material properties from partial observations, which enables them to predict how actions affect future states. As such, embedding such world models in robots allows for better generalization in manipulating a wide variety of garments in the real world. However, learned world models typically require large amounts of real world data that is strenuous to collect, and a mechanism to infer latent properties that govern dynamics. We present a Soft-Object World Model that performs generative full-state estimation and action-conditioned dynamics prediction using a transformer backbone trained on large-scale simulated soft object interactions, with measures to mitigate the sim-to-real gap. By learning to infer latent shape and material cues through in-context learning over a variable-length partial point-cloud history, the model generalizes across robot embodiments, material parameters, garment types and shapes, and manipulation coordination modes without weight updates. The world model is then integrated with a Goal-Conditioned action planner for motion synthesis. We validate the approach in simulation and real world for bimanual garment folding, reporting improvements in reconstruction fidelity, prediction accuracy, and task success.
manipulationsim-to-realworld modelaction-conditioned - arxiv:2610.30983 · cs.CLA Reality Check of Inference-Time Context Augmentation MethodsHaeun Yu, Geert Heyman, Pepa Atanasova, Isabelle Augenstein
Large language models (LLMs) often fail to utilise the context given in the input, resulting in hallucinations or a fallback to parametric knowledge. Context augmentation methods aim to improve model performance by modifying the input, from inserting filler tokens that add computational time to adding relevant content such as elaboration of a context during inference. However, these methods have been evaluated in isolation and only by their performance gains, leaving it unclear how they compare and whether they reliably improve performance. In this paper, we present an evaluation framework spanning five methods that differ in how informative the inserted content is and apply them at the sentence level across seven models spanning multiple sizes on three datasets. We find that accuracy gains are not always genuine improvements but undirected perturbations: the methods correct some predictions while breaking others, particularly when the inserted text allows only for more computational time. Our analysis reveals that methods that bring no additional content primarily change model predictions near the decision boundary. Our findings highlight the need to evaluate the reliability of the accuracy gains before adopting them in practice.
evaluation framework - arxiv:2610.30980 · cs.LGDiagnostic Loss: Asymmetric Supervision for Radiology Report GenerationIgnacy Kolton, Wenxuan Li, Zongwei Zhou, Jarosław B. Ćwikła +3
Automated radiology report generation requires models to do more than produce fluent clinical narratives: they must explicitly state supported diagnoses while avoiding unsupported ones. However, existing medical vision-language models (MedVLMs) are typically trained with token-level maximum likelihood objectives, which prioritize text agreement over diagnostically critical terms. As a result, models can miss key diagnoses and introduce wrong ones. We identify this as an objective misalignment between conventional language modeling and the asymmetric diagnostic requirements of radiology reporting. To encode these requirements, we introduce Diagnostic Loss (DLoss), a differentiable, architecture-agnostic objective that operates directly on decoder token probabilities. DLoss separately aggregates probability mass over vocabularies associated with diagnoses supported by the reference report and with unsupported alternatives. Using asymmetric pooling across token positions in the report, it encourages supported diagnoses to receive sufficient probability in at least one position while suppressing spurious probability spikes for unsupported diagnoses across all positions during training. The loss can be applied to existing 3D MedVLMs without modifying their vision backbones, language decoders, or inference procedures, and introduces no additional inference-time computation. We train and evaluate DLoss on 50,000 abdominal CT scans across several 3D MedVLM architectures on both in-distribution and out-of-distribution report-generation benchmarks. Across these settings, DLoss improves diagnostic faithfulness while maintaining competitive performance on conventional language-generation metrics, showing the value of asymmetric sequence-level supervision for radiology report generation. The code for DLoss is available at https://github.com/gmum/DLoss.
benchmark - arxiv:2610.30970 · cs.LGJoint or Per-Option? Compact Encoders for Closed-Set Decisions from 17 Million to 1 Billion ParametersTardelli Ronan Coelho Stekel
Many language-model pipelines end in a closed-set decision, such as selecting a tool. Prompting a large language model for it is slow and order-sensitive. We ask how small an encoder for such decisions can be and whether it should read the options jointly or separately. In a three-seed study at 32 million parameters, per-option scoring beats joint scoring on a held-out sample of a 37-benchmark decision suite by 5.8 points of chance-corrected skill at matched optimizer steps and by 2.8 at approximately matched training compute; two external benchmarks agree, though not significantly on their unseen tasks. Crossing reader with initialisation shows that the advantage requires a reranker initialisation, whose query-candidate pair input matches per-option reading; from the raw encoder the two readers are indistinguishable. Per-option scoring needs 5 to 13 times more encoder passes: at most twice the median latency, but more than four times the mean. Of six released joint encoders (17 million to 1 billion parameters), the 68- and 150-million-parameter models exceed a 435-million-parameter DeBERTa-v3 zero-shot classifier by 4.9 and 5.9 accuracy points on 209 overlap-audited System1-Arena tasks and are at least on par with it on the 145 tasks unseen in training. The smallest answers a median request in 14 milliseconds on 16 processor threads, against 728 for the classifier. Reordering options changes 12 to 23% of the joint models' answers (per-option scoring: none). Up to 68 million parameters, they lead their size range on a public leaderboard's private, mainly in-distribution tests.
benchmarkleaderboard - arxiv:2610.30969 · cs.ROIORA: Irregular Object Immobilization under Restricted ActuationSang-Hun Han, Changyeon Won, JunGyu Lee, Min-Gyu Park +1
Stable grasping under task-specific disturbances is important for robotic manipulation. However, most dexterous grasp synthesis methods are developed for highly actuated hands in free space, and low-DoF hands may not be able to realize the contact topologies required by these grasps. Meanwhile, support surfaces can provide useful contact forces, but are often treated only as geometric constraints. We present a support-aware grasp synthesis framework for object immobilization with low-DoF coupled hands. Given irregular object geometry, our method represents both hand and object local surface regions as graphs and searches for kinematically feasible matchings between them. We then refine the wrist pose and motor state for geometric contact feasibility and validates the refined grasps through physical rollouts. Starting from the validated grasp pose, an optional task-guided local search explores nearby grasp configurations by varying the wrist pose and motor state, and selects one according to the specified task condition. Experiments on diverse objects show that our method generates various grasp configurations that achieve high grasp success and strong disturbance resistance. Real-world experiments further demonstrate physically executable grasps which enable object immobilization in cooking scenario.
manipulationdexterousgrasp - arxiv:2610.30965 · cs.LGEnergy and Latency First: How Deployment Constraints Reverse Time-Series Model SelectionShruti Bothe, Illyyne Saffar, Ayush K. Varshney
AI models, and time-series models in particular are typically ranked by accuracy and deployed under constraints absent from that ranking. In this paper, we study a real setting where these constraints are explicit and measurable: handover pre- diction in cellular networks, where inference runs on the User Equipment once per 40 ms measurement period and must complete before the network acts. The timing bound is hard: a late prediction is superseded by the measurement it was predicting, while energy is a cumulative budget spent for the life of the device. Neither appears in the objective these models are selected on. We evaluate eight architectures under a protocol that screens for deadline feasibility at the tail rather than the mean, then ranks the feasible models on measured inference energy, count- ing the feature computation that precedes the forward pass and is usually omitted. Gradient-boosted trees on handcrafted temporal features are the most accurate model (F1=0.974) and among the cheapest to run, while the neural architecture an accuracy-only benchmark would rank first (CNN-1D, F1=0.914) costs over 3 times more energy per inference under a realistic duty-cycled deployment pattern. We give three conditions under which this outcome should be expected, so the finding applies beyond telecommunications. Accuracy-only benchmarking does not merely under-report deployment cost. It can rank first a model that cannot be deployed.
benchmark - arxiv:2610.30963 · cs.LGReDence: A Hybrid Agentic and Decision-Model Framework for Rapid, Evidence-Grounded Detection of Modern Social Engineering ScamsJaishnoor Kaur, Josef Horáček, Ian Waskom, Sayak Saha Roy
Detecting online social engineering (SE) scams often requires examining website workflows, verifying claims against external sources, and connecting evidence across multiple pages. Existing detectors provide limited support for these investigations, leaving analysts to manually gather and validate evidence, which inevitably delays detection and subsequent takedown of these attacks. To address this gap, we present ReDence, an open-source hybrid framework for adaptive, evidence-grounded detection across diverse SE scam categories. ReDence combines two locally hosted models with complementary responsibilities: a fine-tuned multimodal agentic Investigator (Qwen3.5-4B) that performs investigative planning, evidence interpretation, and reporting, and a specialized Decider (Tev1) that handles binary and multiple-choice tasks, including browser-action selection, evidence assessment, and decisions about whether to continue an investigation. By delegating discrete decisions to a specialized model, ReDence reduces reliance on lengthy language-model responses, while preserving the reasoning capabilities required for complex investigations. Our evaluation demonstrates that ReDence achieves 96\% accuracy in detecting scams across twelve SE scam categories, substantially outperforming existing detectors while being at least four times faster than comparable agentic models without compromising investigative depth. ReDence also produces structured, evidence-grounded reports that document investigative actions, supporting and contradictory evidence, and collected artifacts, which allow analysts to independently audit detection decisions and utilize the collected evidence to support subsequent mitigation and takedown efforts.
agentic - arxiv:2610.30957 · cs.LGDoubly Robust Distributionally Robust Offline Contextual PricingMin Xu, Xinyi Yin, Yunfan Zhang, Yuxuan Han +2
Offline contextual pricing often relies on logged observational data, but faces challenges from distributional shifts between training and deployment environments. Distributionally robust optimization (DRO) provides a principled approach to off-policy evaluation and learning (OPE/L). However, existing methods are mostly limited to discrete actions. Recent work has explored DRO for continuous treatments using inverse propensity weighting (IPW), while such IPW-based estimators can be sensitive to the convergence rate of propensity score estimates, particularly when estimated nonparametrically, which may lead to larger estimation errors and regret. In this work, we develop a doubly robust (DR) framework for distributionally robust OPE/L in continuous pricing settings. For evaluation, we propose a localized DR estimator that addresses the computational challenges of worst-case expectations by fitting only a small number of regressions, comparable to standard non-robust DR, while achieving semiparametric efficiency under mild product rate conditions. For learning, we leverage the inherent smoothness of demand noise to handle pricing-specific discontinuities in revenue outcomes (e.g., threshold-based purchase decisions), establishing a finite-sample regret bound of $\tilde{\mathcal{O}}_p(T^{-s/(2s+1)})$ for smoothness orders $s=1,2$. This bound improves upon existing regret rates in existing DRO-based off-policy learning (OPL) for continuous treatments. Extensive experiments under various levels of distribution shift validate our proposed framework.
policy evaluation - arxiv:2610.30954 · cs.LGGround-Truth-First Benchmarking of Reinforcement Learning via Converse Optimality: From Smooth, Fully Observed Systems to Non-Smooth or Partially Observed TasksGrégoire Ouerdane, Sinan Ibrahim, Stefan Streif, Pavel Osinenko
Benchmarking in reinforcement learning is usually relative: algorithms are ranked by returns on environments with unknown optima. Distance to optimality remains unobservable, while conclusions may depend on reward design, stochasticity, hyperparameters, environment selection, and simulator details. We develop a ground-truth-first framework based on converse optimality, constructing each task with a certified optimal value function and policy withheld from learners and used only for evaluation. The framework unifies two modes: sensitivity benchmarking samples certified environments from continuous-time generators to assess robustness to task variation, while oracle-referenced comparison uses native discrete-time generators to evaluate policies via optimality gap, regret, and policy error. Continuous-time models support physically interpretable sensitivity studies but introduce dependence on the simulator step ($Δt$) and numerical scheme. Native discrete-time models align the environment, oracle, and learner on the same transition law, eliminating external integration choices as a confounder. We reproduce quantitative results covering optimizer sensitivity and oracle-referenced comparisons on serial planar arms, dynamically extended nonholonomic vehicles, and quadrotor-inspired tasks. The methodology extends beyond smooth affine Gaussian systems: virtual-input pullback handles non-affine raw-action interfaces and finite-covariance non-Gaussian disturbances. A non-smooth converse optimality formulation enables certified benchmarks with kinked value functions and switching surfaces. For partially observed evaluation, an information-state formulation reduces benchmarking to converse optimality on a sufficient statistic and provides an information-penalty reference. Code, fixtures, and diagnostics are available on the public project page.
benchmark - arxiv:2610.30953 · cs.AIMeasuring the Reasoning-Action Coupling in Driving Vision-Language-Action Models through CounterfactualsGabriel Manalu, Saad Wazir, Thao Do, Seongah Kim +3
Driving vision-language-action (VLA) models bring an appealing form of explainability to a safety-critical domain, since a system that can state the reasons for its maneuvers offers the transparency that deployment and public trust in autonomous driving demand. These models read the world into language before they act on it, describing the scene and reasoning over it in text that ranges from chain-of-thought to multi-stage question answering on the way to a trajectory. That framing quietly assumes two separable properties, namely that the trajectory follows the generated reasoning and that the generated reasoning tracks the causes that actually produce the trajectory. To thoroughly investigate these properties, we apply causal, counterfactual interventions to four driving VLAs across different benchmarks. First, we perturb the models' own reasoning and show that altering its meaning produces small trajectory changes relative to blanking the same models' camera inputs, a pattern that persists across benchmarks and model architectures. We then construct scene-level counterfactuals through photorealistic removal of key objects named in the models' reasoning and analyze whether the generated reasoning and trajectories track these changes. Overall, our findings indicate that the reasoning these models present to account for their resulting driving behavior is weakly coupled to that behavior itself, pointing to limits on the explainability these systems are intended to provide.
vision-language-actionbenchmark - arxiv:2610.30949 · cs.CLIQuest-Q1: Advancing Foundation Capabilities for Agentic CLI SystemsA. Jian Yang, An Huang, Bo Dan, Chen Mu +173
We present IQuest-Q1, a sparse Mixture-of-Experts (MoE) agentic foundation model with 320B total parameters (15B activated). IQuest-Q1 is developed through a three-stage process: (1) stable and efficient pre-training, combining Stable SplitNorm, Muon, and muP with a two-stage curriculum whose second stage increases the coding-data mixture, preserving a wide stability margin between safe and optimal learning rates; (2) fine-grained mid-training, expanding the data mixture across domains and formats and progressively extending the context; and (3) multi-scaffold with multi-version post-training, combining SFT, domain-specialized RL, and MOPD across various scaffolds to integrate complementary training signals and strengthen agentic coding capabilities. Furthermore, model merging combines different experts and checkpoints from different stages while mitigating conflicts among their parameter updates. The resulting IQuest-Q1 demonstrates capabilities in both coding and general tasks. We release IQuest-Q1 as an open-source model to support community research and development.
agenticpost-training - arxiv:2610.30947 · cs.AIRetiJev: Fast and Source-Grounded Ophthalmic Decision ModelsBo Liu, Lianyu Wang, Meng Wang, Ching-Yu Cheng +1
Ophthalmic clinical decision support requires fast, accurate, and privacy-preserving inference, yet autoregressive large language models (LLMs) incur substantial latency and computational overhead for decisions among predefined actions. We therefore investigate a decision-oriented alternative to generative reasoning. Specifically, we first introduce RetiJev-Data, a source-grounded ophthalmology corpus comprising 78,158 clinical-decision and diagnosis questions derived from articles in PubMed Central and a MeSH vocabulary covering 531 eye diseases, and 50,074 guideline-grading questions spanning 11 grading systems, along with a 24,662-question evaluation benchmark. Based on this, we develop RetiJev, a family of 0.8B-9B decision models that directly score candidate answers without autoregressive decoding. Across all model sizes, RetiJev outperforms both a 27B LLM and the commercial Jev 1.13 decision model on guideline grading by 10.4-23.0 points. RetiJev-9B further achieves the highest accuracy on clinical decisions, improving over both baselines by 5.4-6.5 points while matching them on diagnosis. It answers a case in 158 ms on one GPU, 3.5x faster than LLM generation and 470x faster than reasoning. Models are released at https://github.com/Awenbocc/RetiJev.
benchmark - arxiv:2610.30945 · cs.AIMetacognitive Scaffolds Create Productive Friction for AI-Assisted Knowledge WorkViktor Kewenig, Lev Tankelevitch, Sean Rintel, Advait Sarkar +1
Generative AI imposes new metacognitive demands on knowledge workers as their work shifts from production to planning, evaluation, and orchestration. In a randomized, between-subjects experiment on AI-assisted writing (297 workers, 594 essays), we test dynamically-generated metacognitive scaffolds supporting goal specification and task decomposition. We find that scaffolds improve writing quality, with this boost persisting in short-lag transfer when scaffolds are removed. We also find evidence of improved topic recall three days later. These effects are consistent with observed changes in how people learn to prompt AI systems after exposure to scaffolding. However, scaffolded participants rate the AI as less helpful and more cognitively demanding even as they perform better, suggesting that subjective ease can understate the value of support. Our findings make the case for metacognitive scaffolds in AI-assisted knowledge work, revealing a productive friction worth designing for as workers increasingly delegate work to agentic AI.
agentic - arxiv:2610.30942 · cs.LGGateFormer: Forecast-Space Decomposition for Intra-Hour Solar Irradiance Prediction from Sky ImagesDavid Hamlyn, Sunny Chaudhary, Jonathon Hare, Tasmiat Rahman
Sub-hourly dispatch, balancing, and storage control increasingly depend on forecasts that anticipate rapid variability in solar generation. Accurate sky-image-based irradiance forecasting requires a combination of visual cloud dynamics with recent irradiance state while adapting to different short-term irradiance-change regimes. To address this, we introduce GateFormer, a GHI-conditioned dual-decoder ViT for intra-hour global horizontal irradiance (GHI) forecasting through forecast-space decomposition. GateFormer embeds observed GHI into the visual token stream, allowing interaction with sky-image patch features through self-attention. From this shared representation, a Trajectory Head predicts future GHI sequence, while a Delta Head predicts final-horizon change from the latest observed GHI. The two final-horizon forecasts are adaptively weighted in prediction space, with magnitude- and direction-supervision targeting the remaining error. On the NREL-TSI benchmark, GateFormer attains 125.01 W/m^2 RMSE at +30 min, improving over a state-of-the-art transformer baseline by 9.6% under paired-timestamp evaluation. Across NREL-TSI, Chilbolton, and SIRTA, GateFormer achieves the lowest final-horizon RMSE among the evaluated baselines, with forecast skill of 22.97-28.86% relative to smart persistence. Component ablations show adaptive weighting outperforms both individual decoders and fixed averaging, distinguishing GateFormer from conventional forecast combination, while the supervised correction further reduces RMSE by 1.68 W/m^2 and MAE by 4.86 W/m^2. Ramp-regime diagnostics show larger gains during operationally important positive ramps, reducing RMSE by 26.27 W/m^2. These results identify forecast-space decomposition as an alternative forecasting formulation, with potential relevance to other multimodal short-horizon forecasting tasks and deployment using only sky images and measured GHI.
benchmark - arxiv:2610.30933 · cs.CLReflectPO: Reflection-Guided Policy Optimization for Long-Horizon LLM AgentsRumei Li, Xunliang Cai
With external tools, large language models (LLMs) can tackle complex tasks through sustained environmental interaction. LLM-based agents are therefore increasingly used to tackle complex long-horizon tasks in real-world environments. However, these tasks involve sequential, interdependent decisions and changes in environmental states. Early execution errors affect subsequent interaction histories and environmental states, and their effects accumulate along the decision chain. To address this challenge, we propose Reflection-Guided Policy Optimization (ReflectPO), a reinforcement learning framework for autonomous error recovery in long-horizon tasks. ReflectPO comprises two core components. The first is Reflection-Guided Sampling. We use branch differences and repeated visits in interaction trajectories to construct hints for error recovery. The second is Tree-Structured Credit Assignment. We organize sampled trajectories into trees and assign action-level credit using parent--child reward differences and recovery gains. Experiments over two long-horizon tool-use benchmarks show that ReflectPO outperforms multiple existing baselines in most domains. It also achieves the best overall performance on average. Further analyses show that ReflectPO effectively improves agents' ability to recognize their own deviations, recover from erroneous states, and continue making progress toward task completion.
llm agenttool-usebenchmark - arxiv:2610.30928 · cs.CVICGF-CXRMate: Image-Conditioned Clinical Context Compression and Gated Fusion for Chest X-ray Report GenerationVu Minh Tran, Thuc Huynh, Son Thanh Luu, Doanh C. Bui
Auxiliary patient data can improve chest X-ray report generation, but directly concatenating heterogeneous clinical information with visual tokens produces long, variable-length decoder prompts and may expose the model to redundant or image-irrelevant context. We propose ICGF-CXRMate, an image-conditioned clinical context compression and gated fusion framework. Learnable queries are first conditioned on the current chest X-ray representations and then attend to the patient-data sequence, compressing it into a fixed number of clinical context tokens. Gated cross-attention selectively integrates these tokens into the visual stream, while the compressed tokens are retained in the decoder prompt as a direct clinical pathway. Experiments on linked MIMIC-CXR and MIMIC-IV-ED examinations show that ICGF-CXRMate improves report quality while reducing the average prompt length. In Stage~2, it improves RadGraph-F1 from 24.00 to 28.86, CXR-BERT similarity from 56.86 to 62.80, and BLEU-4 from 5.14 to 7.96 compared with CXRMate-ED, while reducing the average prompt length from 373.89 to 288.37 tokens, corresponding to a 22.9\% reduction. After self-critical sequence training, ICGF-CXRMate outperforms CXRMate-ED on six of eight report-quality metrics in Stage~3. Ablation studies confirm the complementary contributions of image-conditioned compression, gated fusion, and direct retention of the compressed clinical tokens.
context compression - arxiv:2610.30920 · cs.AIDConstrain: Compiling Context-Parameterized Transitions for Diffusion Language Model Constrained DecodingShidong Cao, Jing Ma, Wenhao Chai, Yuchen Tian +2
Diffusion language models (DLMs) decode multiple tokens in parallel, so individually valid tokens can be incompatible together under a schema. Exact joint inference over a grammar graph enforces compatibility, but it requires token-labeled successors across many hypothetical parser configurations. Encoding stacks and field histories in compiled states can cause exponential growth, which forces prior exact decoders to fix property orders and exclude valid outputs. We present DConstrain, which compiles schemas into context-parameterized transition programs. Cold compilation specializes token consumption to local grammar states while retaining context-dependent stack operations, environment guards, and caller-dependent deferred parsing symbolically. Tracking field histories as runtime values supports arbitrary property order without enumerating field subsets offline. At runtime, warm compilation binds reached configurations to these programs to construct exact successors, overlapping with the model forward pass because graph construction requires no model scores. Furthermore, token-effect grouping merges tokens with identical compiled effects before runtime contexts are bound, so groups are reused across stacks and environments even when concrete successors differ. Objective-specific score aggregation and token restoration provably preserve both the Viterbi optimum and the token-labeled path distribution of ungrouped inference. Across three benchmarks and two DLM families, DConstrain compiles 100% of BFCL schemas under arbitrary property orders, whereas an optimized finite-state baseline compiles only 71.30-85.81%. Under arbitrary property order, it reduces BFCL end-to-end decoding latency by 6.65-46.26 times with 100% schema validity and raises DiffusionGemma's BFCL accuracy from 37.27% under fixed order to 81.56%.
benchmark - arxiv:2610.30917 · cs.AIHarnessing Language Models as Optimization Modeling AgentsYunxiang Luo, Xin Xu, Tianhao Chen, Zhixi Xiong +6
Large language models (LLMs) can translate natural-language optimization problems into solver programs, but reliably producing correct formulations and implementations remains challenging. Existing approaches improve optimization modeling through agentic workflows or task-specific post-training. This raises a practical question: can a lightweight harness using a single model role achieve competitive performance without additional training? We introduce OPTICYCLE and OPTISTATE, two lightweight inference harnesses for optimization modeling, and evaluate them on nine benchmarks comprising 2,634 problems using 11 Qwen variants spanning 0.6B-27B parameters. Our results show that execution-guided revision consistently improves macro accuracy over direct generation for all evaluated models, with gains of up to 49.46 percentage points. Despite their simplicity, both harnesses outperform six adapted agentic workflows in same-model comparisons and four evaluated post-trained optimization models at comparable model sizes, without additional training. Explicit formulation state provides further but model-dependent improvements, suggesting that additional structure is not universally beneficial.
agenticpost-trainingbenchmark - arxiv:2610.30914 · cs.LGMIG-COW: Transferable Adversarial Attacks on Deepfake Detectors via Gradient DecompositionWonjune Seo, Joonhyuk Baek, Yeseong Jung, Saerom Park
Despite recent advances, deepfake detectors remain vulnerable to adversarial examples, particularly in diverse, real-world settings. We propose MIG-COW, a novel adversarial attack framework that generates highly generalizable and visually imperceptible adversarial examples. By combining momentum-integrated gradients with a consensus-orthogonal decomposition, MIG-COW captures both shared and model-specific vulnerabilities across heterogeneous CNN and ViT detectors. On the AADD-2025 Challenge benchmarks, MIG-COW achieves a 99.96% white-box attack success rate (ASR) with high perceptual similarity (SSIM), significantly outperforming existing baselines. However, its limited 7.16% ASR against official black-box targets - despite achieving the best overall score - highlights the ongoing challenge of transferability. We also demonstrate that incorporating low-performing but diverse models in the ensemble can degrade attack effectiveness, underscoring the need for careful surrogate model selection in real-world adversarial settings.
benchmark - arxiv:2610.30903 · cs.LGSame molecule, Different Knowledge: Cross-Identifier Binding Failures in LLMs For chemistryKhouloud Saadi, Di Wang
Chemical AI workflows routinely represent the same molecule through multiple co-referring identifiers, including trivial names, IUPAC names, and SMILES, implicitly assuming that these representations provide interchangeable access to an LLM's knowledge. To test this assumption, we first introduce a controlled benchmark and show that it does not hold. Across foundation models spanning 7B to 72B parameters, even chemistry-specialized models, co-referring molecular identifiers consistently expose different subsets of the model's knowledge. We then conduct a series of controlled behavioral, representational, and causal analyses to characterize this failure. We show that even when a model can identify a molecule from its SMILES and retrieve a fact through the corresponding name, the same fact can remain inaccessible directly from SMILES. We term this phenomenon \emph{cross-identifier binding failure}. Further analysis of internal representations, conflicting identifiers, and parameter editing shows that neither improved chemical representation competence nor modification of individual facts is sufficient to establish reliable access across identifiers. Together, these results indicate that the core failure lies in the binding between an identifier and knowledge already present in the model. Motivated by this diagnosis, we introduce Binding Self-Distillation, a simple mitigation that uses lexical identity during training to transfer the model's existing knowledge access to the structural route. This enables factual retrieval directly from SMILES at inference time, without requiring a name lookup or training the underlying model from scratch.
benchmark - arxiv:2610.30890 · cs.LGContinuous-depth Deep Gaussian ProcessesYing Li, Sanyou Wu, Zi Yang, Qiaochu Xu +5
Deep Gaussian processes (DGPs) are appealing Bayesian regression models, but their standard discrete-depth parameterization often becomes harder to optimize as depth grows. We argue that depth is better treated as a latent continuous evolution than as a long discrete stack. We therefore propose a continuous-depth formulation in which the latent representation follows a stochastic differential equation with a depth-indexed drift family equipped with a GP-based prior. This reframes the core design choice as the prior placed on that family, rather than adopting a purely neural continuous-depth parameterization as in neural ODE or neural SDE models. We study two instantiations of this idea: CDGP, a direct GP prior over the state--depth drift, and FlowDGP, a flow-evolved prior that captures richer depth dependence at lower training cost. To motivate the redesign, we analyze how the single-sample DGP objective can become increasingly sensitive under repeated layer composition, while residual DGP is a useful first repair that still remains a finite discrete construction. We also give pathwise sensitivity analysis for the continuous-depth formulation. Experiments on synthetic, benchmark, and large-scale regression tasks show that residual DGP improves over plain DGP, while CDGP and FlowDGP provide a stronger overall story across the regimes we study.
benchmark - arxiv:2610.30888 · cs.LGBuilding Presentation Intelligence for High-Quality Slide GenerationYuhang Xie, Cheng Li, Chenliang Li, Mingyu Xu +5
PowerPoint presentations (PPTs) are a dominant medium for communication across diverse domains, yet high-quality PPT generation still depends heavily on labor-intensive human effort. Recent advances in large language models (LLMs) for agentic tasks have created both new opportunities and challenges (Chen et al., 2026) for automatic PPT generation. In this work, we present a unified training framework for automatic PPT generation toward presentation intelligence, integrating cross-validated data synthesis, guided teacher enhancement, and verifiable reward optimization. We first construct a cross-validated pipeline for data synthesis, with explicit control over both difficulty and validity. We then develop a teacher-enhancement framework grounded in rules and prior knowledge to substantially strengthen the instructional capacity of teacher models. Building on this, we further improve supervised fine-tuning (SFT) through a mixed-teacher strategy and a lightweight outcome-driven trajectory filtering method. Finally, based on the resulting SFT models, we introduce a suite of verifiable aesthetic metrics together with a GRPO-based reinforcement learning framework to optimize slide-level visual attributes beyond text-only supervision. Experiments show that our enhanced SFT pipeline improves Qwen-3.6-27B and Qwen-3.6-35B-A3B by 7.45 and 9.35 points on PresentBench, respectively, surpassing GLM-5.2 while maintaining strong generalization. Reinforcement learning further improves performance across five aesthetic dimensions, compensating for the limitations of SFT in visual refinement. Overall, the results demonstrate the effectiveness of our framework for high-quality PPT generation.
agentic - arxiv:2610.30885 · cs.CLFrom Perpetrators' Digital Footprints to Behavioral Communities: An Ontology-Driven Framework for Bengali Cyberbullying Analysis--- FULL VERSIONMd. Nesarul Hoque, Md Hanif Seddiqui, Miskatul Anwara, Rudra Pratap Deb Nath
Cyberbullying involves behavioral and contextual patterns beyond isolated abusive messages. However, existing approaches remain fragmented across content detection, semantic representation, and network analysis, with limited interpretable perpetrator representations, multi-scale structural modeling, and explanations of behavior-driven similarity and influence propagation. These limitations hinder the identification of persistent perpetrators, characterization of behavioral communities, and assessment of potentially influential users. This paper proposes Ontology-Driven Cyberbullying Behavior and Network Analysis (ODCBNA), a perpetrator-centric framework integrating requirements-driven ontology design, knowledge graph construction, behavioral modeling, community analysis, and influence modeling. Using the Bengali Contextual and Diverse Cyberbullying (BenCoDiC) corpus as a use case, ODCBNA identifies $1{,}063$ users with at least three bullying comments or replies as perpetrators and represents each through a $64$-dimensional behavioral and contextual feature vector. Cosine similarity and $k$-nearest-neighbor modeling produce a weighted network with $27{,}554$ edges. Louvain analysis identifies six communities with a modularity of approximately $0.495$ and $27$ sub-communities. Feature-contribution analysis identifies the Video, Impact, and User Profile groups as the strongest contributors to behavioral similarity and community separation, while contextual dimensions support interpretation. Community-constrained influence analysis shows that IM Greedy achieves small positive gains in all six communities, although different seed-selection methods produce comparable diffusion spreads despite limited seed overlap. These findings provide an interpretable basis for understanding perpetrator behavior, community organization, and potential influence patterns in Bengali online environments.
knowledge graph - arxiv:2610.30884 · cs.LGTracing Subliminal Learning Through PretrainingAngela Tang, Vishesh Anand, Gonçalo Paulo
Large language models trained on data generated by other models can sometimes inherit traits unrelated to that data, a phenomenon called subliminal learning. We study both hard label distillation, where students train on numbers generated by a biased teacher, and on-policy soft label distillation, where students generate the data and the teacher's logits provide the learning signal. Using Pythia, PolyPythia, and DataDecide OLMo models, we find that transfer persists in both settings across differences in weight initialization, pretraining data order, pretraining corpus, model size, and architecture. Across pretraining checkpoints, the ability to transmit a preference and the ability to acquire one develop separately: some checkpoints are more effective teachers but weaker students, and vice versa. Consistent with this distinction, cross-model transfer can be strongly asymmetric in magnitude when teacher and student roles are reversed. To investigate what carries the preference signal, we construct synthetic number sequences that match the teacher's generated number frequencies but have no dependence on the prompt or preceding numbers and find that these still induce substantial transfer in the hard-label setting. In top-$K$ soft-label distillation, removing animal tokens from the loss reduces transfer, although not completely in the same-model setting. In our experiments, soft label distillation produces more consistent transfer with less extensive training than hard label distillation. The growing use of on-policy distillation in model post-training makes this a particularly relevant setting for assessing unintended preference transfer.
post-training - arxiv:2610.30880 · cs.LGOne Teacher, Many Views: Reducing Reference Sensitivity in On-Policy Self-DistillationKhouloud Saadi, Di Wang
On-policy self-distillation (OPSD) improves reasoning by training a model on its own trajectories with dense token-level supervision from a privileged copy conditioned on a reference solution. Existing methods typically use a single teacher distribution conditioned on the complete reference. We identify a limitation of this formulation: a reference is only one realization of the reasoning process. Although it provides evidence about the answer, it also encodes an ordering, phrasing, and solution path. Full-reference conditioning can therefore produce correct targets that depend excessively on information unavailable to the student at inference. We show that replacing one valid reference with another, while holding the problem and student state fixed, can substantially alter the teacher distribution, a phenomenon we call reference-induced target instability. To address this instability, we introduce Monte Carlo Privileged Distillation (MCPD), which models privileged context as a distribution over partial reference views and distills the marginal teacher distribution. MCPD evaluates the teacher under multiple views and averages the resulting next-token distributions. We show that this probability-space average is the exact target induced by minimizing the expected teacher-to-student KL divergence over privileged views. Across mathematical and scientific reasoning benchmarks, we show that MCPD produces more reference-consistent supervision and consistently improves over full-reference OPSD.
benchmark - arxiv:2610.30876 · cs.ROLocoManiLab: A Unified Testbed for Data, Policies, Controllers, and Deployment in Humanoid Loco-ManipulationDewei Wang, Jinrui Han, Jiyuan Shi, Chenyun Zhang +7
Humanoid robots must coordinate locomotion and manipulation in everyday environments, yet progress is difficult to assess across differing policies, controllers, and task settings. LocoManiLab provides an integrated testbed spanning data collection and generation, policy training, evaluation, and physical deployment. Its task suite covers 234 atomic and compositional tasks across five household scene categories, supported by 51.5 hours of simulated teleoperation. The evaluation covers twelve policies with a shared controller across task and scene splits and perturbation levels, two policies across five controllers, and studies of demonstration scale and data-source mixtures. Success rates fall on unseen tasks and scenes and generally decline as visual perturbations increase, with differences in generalization and robustness across policies. For data generation, a humanoid motion generation model is fine-tuned for loco-manipulation to study the alignment of its outputs with task execution requirements and its integration into the data generation pipeline. Three simulated loco-manipulation tasks closely match real-world setups, enabling evaluation of three policies in both domains. Similar policy rankings support the reliability of simulation-based evaluation.
manipulationhumanoidteleoperation - arxiv:2610.30867 · cs.AISeparate Field Scoring Protects Unused Tools When Agents Pool Tool-Use ExperienceAbhijit Chakraborty
Agents that call tools from a shared catalog can pool what they learn about which tool serves which request. We measure what pooled usage examples do to tool retrieval, reporting separately the held-out tools, for which no agent has usage, and all other tools. Adding usage examples to the index as extra entries for used tools raises Hit@5 (whether a gold tool is in the top five) for those tools and lowers it for held-out tools, by a large margin on two of three catalogs and with both embedders we tested. Giving used tools the same number of equally long entries filled with other tools' text lowers held-out Hit@5 about as much, but gives little or none of the gain. We call our method of scoring usage as its own field SYNAPSE. On StableToolBench, the held-out gold tool appears in its top five more often than with a router that reserves slots for usage (0.785 against 0.683), and this held-out lead holds on all three catalogs with both embedders. This protection persisted when we replaced each tool's usage with other tools' text, and a constant score offset for tools with usage gives similar held-out Hit@5 in most settings. The gain for tools with usage, in contrast, needs usage about the right tool. The router keeps more of the gain on labeled queries on TOUCAN with both embedders and on ToolRet with one, so the better choice depends on the workload. Splitting the usage among ten simulated agents shows that pooling helps exactly as much as giving one agent the same records.
agenttool-use - arxiv:2610.30866 · cs.CVVerus-Insight: Towards Explainable Artifact Detection in AI-Generated VideosZhuoran Zhang, Yang Shi, Qixun Wang, Tengfei Liu +10
Recent advances in video generation have made synthetic videos increasingly realistic, placing greater demands on reliable AI-generated video detection. Existing artifact-based detectors typically identify generation artifacts as evidence for source prediction, yet they may either fail to recognize subtle artifacts or reach the correct source decision without faithfully identifying the evidence underlying it. In this work, we systematically investigate how to design and train effective artifact detectors and formulate artifact detection as a structured process comprising three complementary capabilities: Inspection Planning, which determines what should be examined; Evidence Verification, which verifies whether the inspected content exhibits genuine generation artifacts; and Evidence Aggregation, which integrates the verified evidence into the final source decision. Building on this formulation, we introduce Verus-Insight, trained with human-anchored full-process supervision and progressive capability training to strengthen both fine-grained artifact detection and video-level detection. We further introduce Verus-Bench, a benchmark with fine-grained human annotations of perceptible artifacts for evaluating artifact detection. Extensive experiments across diverse benchmarks demonstrate that Verus-Insight achieves strong performance in both fine-grained artifact analysis and AI-generated video detection, with substantial gains in recovering human-perceived artifacts. Further analyses reveal that the three capabilities are complementary yet distinct, and that Verus-Insight more reliably converts relevant inspection targets into supported artifact evidence without increasing spurious artifact claims.
benchmark - arxiv:2610.30864 · cs.MAConsensus-Based Distributed Filtering for Angles-Only Tracking in Cislunar SpaceGiovanni Romagnoli, Fabrizio Pilone, Giordana Bucchioni
This paper investigates consensus-based distributed filtering strategies for multi-observer angles-only navigation in the cislunar space. The objective is to estimate the inertial state of a target spacecraft using only line-of-sight measurements collected by a network of observer spacecraft. This problem is particularly relevant for autonomous cislunar navigation, where ground support may be limited, and centralized fusion architectures may be problematic due to communication and computational constraints, and single-point failure risks. Three distributed filtering approaches are considered: the Distributed Extended Kalman Filter, the Information-Weighted Consensus Filter, and the Kalman Consensus Filter. All methods share the same local prediction and measurement-processing steps, but differ in the way information is exchanged and fused across the network. Their performance is evaluated and compared to a centralized Extended Kalman Filter, adopted as benchmark, and independent local filters in a high-fidelity cislunar scenario. Numerical simulations show that the distributed approaches significantly improve estimation accuracy with respect to independent filters. Monte Carlo sensitivity analyses are then conducted to characterize the robustness of the considered methods to finite consensus iterations and prior misalignment among the agents.
benchmark - arxiv:2610.30860 · cs.CLSynthetic-to-Real Code-Switched ASR: A Spontaneous Nepali-English Podcast BenchmarkFiroj Paudel
Nepali-English code-switched conversational speech presents severe challenges for automatic speech recognition (ASR), chiefly through script confusion (inappropriate mixing of Devanagari and Latin orthography) and elevated error rates at language switch points. Supervised fine-tuning (SFT) can adapt multilingual foundation models, but human code-switched audio is expensive to collect and annotate. We curate a new benchmark of 13,044 spontaneous podcast utterances (15.45 h total; 10,214 test, 2,830 dev), audited and verified to eliminate machine translation leakages and acoustic loops. We investigate a two-stage synthetic-to-real adaptation framework: Stage 1 applies SFT on 0.95 h of speech synthesized via a multi-speaker, dual-engine TTS pipeline; Stage 2 explores Group Relative Policy Optimization (GRPO) with transcription and script rewards. We find that synthetic SFT provides the primary cross-domain transfer to spontaneous human speech, reducing Mixed Error Rate (MER) from 23.40% to 21.52% (-1.88 pp) and switch-point error rate (SPER) from 35.89% to 28.77% (-7.12 pp). Subsequent GRPO yields small, statistically significant adjustments on 173 test utterances (1.69%), driving English script confusion (SCR_en) from 6.45% to 6.36% (p <= 0.001) and SPER to 28.70% (p = 0.022 clustered). Crucially, ablation reveals that in this SFT-converged encoder-decoder setting, an auxiliary script reward term yields performance identical to MER-only optimization because rollout candidate groups exhibit near-zero script variance on synthetic data. On monolingual benchmarks, SFT improves Nepali MER by 2.71 pp (stabilizing at 18.42% after GRPO, a 2.38 pp gain over baseline) while preserving English lexical WER (3.38% vs. 3.33%). Verified transcripts, timestamps, episode metadata, and reconstruction scripts are publicly released.
benchmark - arxiv:2610.30854 · cs.CVCAVC: Convergence-Adaptive Audio-Visual Cache for Modality-Asymmetric Diffusion GenerationLifan Jiang, Yibo Zhao, Senlin cheng, Bolei Xu +5
Diffusion-based audio-visual generation methods typically perform joint denoising of audio and video latents throughout the entire sampling trajectory, repeatedly executing modality-specific denoising branches and cross-modal attention at every timestep. However, we observe that such uniform joint computation is not always necessary. The key is a modality-asymmetric recoverability: audio-side drift introduced by mid-stage caching can be substantially reduced by only a few terminal joint denoising steps, whereas video-side degradation caused by caching leads to persistent loss of motion and temporal details. This asymmetry suggests that inference acceleration should selectively cache recoverable audio-side computation while preserving full video refinement. Motivated by this observation, we propose \textbf{Convergence-Adaptive Audio-Visual Cache (CAVC)}, a training-free modality-asymmetric caching framework. CAVC adopts a sandwich-structured computation allocation: after detecting that audio denoiser outputs enter a cache-tolerant regime, it caches and reuses the audio denoiser output to bypass audio denoising and cross-modal attention in the middle stage, while the video branch continues independent spatio-temporal refinement; a few joint steps are then restored at the end to reduce accumulated cross-modal drift and improve synchronization. Experiments across multiple public benchmarks demonstrate that CAVC achieves 1.10--1.35x wall-clock acceleration, improves several visual dynamics and temporal statistics, and maintains competitive audio fidelity and audio-visual alignment.
benchmark - arxiv:2610.30853 · cs.LGTraining Mixed-Monotone Neural Networks for Verification Without Set PropagationJonas Fingerhut, Lukas Koller, Matthias Althoff
Most formal verification approaches for neural networks conservatively enclose the output set of the neural network. However, computing a tight output enclosure is in general computationally hard. We sidestep this challenge with a novel neural network architecture that realizes mixed-monotone functions as the difference of monotone subnetworks. By utilizing the theory on mixed-monotone systems, this reduces the computation of a tight output enclosure from complicated set propagation to two simple forward passes of opposite corners of an input interval. Moreover, because the size of the resulting enclosure is a closed-form function of the weights of the neural network, we can directly minimize it during training. We prove a lower bound on the size of any such enclosure, given by the total variation of the represented function generalized to several inputs, which makes it possible to compare the tightness of the enclosure against the optimum. Our evaluation on regression and dynamical-system benchmarks demonstrates that our training reduces the enclosure size to within 0.5% of the lower bound without sacrificing prediction accuracy. Computing our enclosure is orders of magnitude faster than CROWN and $α$-CROWN and on deep networks it is also tighter than zonotope propagation and CROWN, while its size exceeds that of $α$-CROWN by only 5.1%.
benchmark - arxiv:2610.30851 · cs.AIADAPT for Science: A Multi-Agent System for Self-Evolving Scientific OptimizationZhaoran Feng, Yuchen Ni, Jianan Liu, Lemao Liu +2
Scientific discovery often requires identifying a small number of high-performing designs in vast, constrained search spaces. This is difficult when evaluations are costly, feasible solutions are sparse, parameters are strongly coupled, and physical constraints are strict. Recent AI for AI methods can automatically improve algorithms and optimization procedures. However, general purpose approaches are not designed around scientific priors, physical constraints, and limited evaluation budgets. We introduce ADAPT for Science, a multi-agent framework that treats scientific optimization as two coupled decisions: where to search and how to search. ADAPT uses historical designs to identify promising and diverse initial conditions, progressively concentrates computation on favorable search directions, and evolves the optimization procedure from execution feedback. We evaluate ADAPT on the design of gravitational wave detectors using a differentiable physics simulator and eight unseen detector topologies. Under matched scientific evaluation budgets, ADAPT achieves a mean loss of 0.204 +/- 0.101, reducing loss by 54.9% relative to the best classical optimizer and by 62.2% relative to the best general purpose agentic evolution baseline. ADAPT achieves the lowest loss on all unseen topologies among the evaluated baselines. ADAPT provides a step toward scientific optimization systems that continuously learn from prior experience and their own search outcomes.
multi-agentagenticagent frameworkagent systemself-evolving - arxiv:2610.30843 · cs.LGLean Finder V2: Semantic Search for Mathlib That Disambiguates User QueriesJialin Lu, Xiaoyu Liu, Hengrui Luo, Rodrigo Stehling +4
We introduce Lean Finder V2, a semantic search framework for Mathlib that disambiguates user queries through complementary model-level adversarial training and input-level query rewriting. Human mathematicians and coding agents rarely search Mathlib by providing detailed context. In practice, they query short properties, notation, aliases, or hypotheses, leaving much of the intended mathematics implicit. In contrast, existing Mathlib search engines unrealistically assume context-rich queries. The scarcity of real-world short queries paired with ground-truth Mathlib statements makes this ambiguity even more intractable. We argue that this is essentially a data problem. Through adversarial training, we scale up the generation of labeled short and ambiguous queries, and learn to disambiguate user queries at both model-level fine-tuning and input-level query rewriting. On the human-curated MathlibQR benchmark, our Lean Finder V2 sets a new state of the art, with at least 11% relative improvement over the previous best Mathlib semantic search engines. On top of our state-of-the-art retriever, optimized query rewriting further improves MathlibQR retrieval. On unlabeled queries from human and coding-agent interactions, Lean Finder V2 is ranked as the strongest system. Integrated with Claude Code, Lean Finder V2 improves proof success on downstream theorem-proving task. Lean Finder V2 is available at: https://leanfinder.github.io/
benchmark - arxiv:2610.30841 · cs.CLTRIAD: Benchmarking Omni-Modal Ambiguity in Multimodal Large Language ModelsZhaolu Kang, Yidi Wang, Yile Li, Siqi Zeng +3
Ambiguity is a central feature of natural communication: a sentence, image, or sound may remain underspecified until evidence from the other channels is considered. Yet current multimodal benchmarks mostly test perception, recognition, or reasoning over already determinate inputs, leaving unclear whether omni-modal large language models can resolve ambiguity distributed across text, vision, and audio. We introduce TRIAD, a benchmark for omni-modal ambiguity resolution. Each item is a text--image--audio triplet paired with a question and a set of answer options, and is constructed so that the full triplet determines a unique answer while removing any one modality makes the set of consistent options non-unique. TRIAD contains $327$ hand-written queries in $86$ scene groups and covers $18$ ambiguity categories across the three modalities. Evaluating $14$ omni-modal systems reveals that strong item-level performance does not translate into robust scene-level disambiguation: the best model still trails humans by over $20$ points in group accuracy. Removing answer options widens the gap further, driving model group accuracy to the single digits while humans remain robust. Leave-one-out and transcript/description counterfactuals further show that current systems often underuse audio cues when text and image are present. TRIAD reframes omni-modal evaluation around disambiguation rather than recognition, and provides a taxonomy, benchmark, and diagnostic protocol for measuring this capability.
benchmark - arxiv:2610.30834 · cs.CVRECAP: Looking Once Is Not Enough for Vision-Language ReasoningZhaolu Kang, Tailong Luo, Chenxin Li, Yingjie He +12
Long chain-of-thought reasoning improves vision-language models (VLMs), but it also exposes a temporal grounding failure: as generation unfolds, models may drift from image evidence while reinforcement learning still provides only a final-answer reward. We study this gap through visual revisit, a temporally localized reactivation of image evidence during reasoning. Across three VLM backbones, revisit peaks are predictive of correctness and causally linked to performance: masking them degrades accuracy more than masking non-peaks, while injecting revisit-like peaks into failed rollouts partially restores correct answers. Control analyses show that this signal is not explained by raw attention strength, uncertainty, logit margin, or hidden-state norm. We propose RECAP, a GRPO-compatible training method that turns detached revisit traces into a credit-assignment signal. RECAP uses revisit-conditioned advantage estimation, dynamic discounting, and gated value prediction to propagate final rewards through visually important reasoning steps, without extra rewards, additional rollouts, or inference-time changes. Across Qwen2.5-VL-7B, InternVL3-8B, and LLaVA-OV-7B, RECAP consistently improves over GRPO and VPPO, yielding $+6.1$--$+6.5$ HallusionBench gains over the SFT base while preserving reasoning and text-only performance. Results on larger LoRA-tuned models, Chinese zero-shot benchmarks, visual perturbations, and human evaluation further demonstrate robust improvements in image-grounded reasoning.
benchmark - arxiv:2610.30828 · cs.CLTutorCaseBench: A Case-Grounded Benchmark for Evaluating LLM Tutors in Supporting Students with Challenging BehaviorsYeon Su Park, Chanyoung Park, Tae Soo Kim, Seungju Kim +4
LLMs are increasingly used as AI tutors, yet existing benchmarks overlook the challenging behaviors common among students with learning and behavioral difficulties. We introduce TutorCaseBench, a benchmark that grounds AI-tutor evaluation in qualitative case studies from pedagogy, educational psychology, and special education. Our pipeline converts documented student categories into case-specific scenarios, tutor-student dialogues, and rubrics grounded in source evidence. The benchmark contains 923 dialogues across five challenge types, evaluated through open-ended response generation (1,560 items) and multiple-choice selection (2,537 items). Across 11 state-of-the-art LLMs, even the strongest model satisfies a rubric criterion in only 55.8% of open-ended cases, while models are better at selecting appropriate strategies than generating situated responses. These results reveal a gap between intervention recognition and context-specific adaptation, especially for open-weight models. All data are publicly available at https://github.com/kixlab/TutorCaseBench.
benchmark - arxiv:2610.30827 · cs.CVOne Evaluation Is Enough: Inversion-Free Image Editing with Trajectory CorrectionLifan Jiang, Taiyang Chen, Senlin cheng, Bolei Xu +5
Real-image editing commonly relies on inversion and iterative sampling, while efficient one-step editors often stabilize a large update by estimating and averaging editing fields at multiple noise levels. We ask a stricter question: is a single temporal field evaluation sufficient for reliable inversion-free editing? Our answer rests on a previously underused duality in a variance-preserving (VP) one-step model: one source-target query at a shared noisy state simultaneously determines a local angular semantic tangent and a clean endpoint displacement. We integrate the tangent over the analytically known remaining VP angle to obtain a nominal one-step transport, and use the endpoint displacement to correct its long-horizon extrapolation through a constrained projection. This transport-first, correction-second construction yields Endpoint-Calibrated Transport (ECT), a closed-form, training-free editor requiring only one noise level, two text conditions packed into one denoiser batch, and one latent update. ECT requires no image inversion, trajectory integration, second temporal observation, or iterative optimization, and can be directly applied to pretrained epsilon-prediction VP one-step generators. Across the reported benchmark settings, ECT delivers stable gains in editing quality within an extremely short editing time.
benchmark - arxiv:2610.30824 · cs.CVLoop to Look Better: Learning Adaptive Depth for Question-Guided Perception in Vision-Language ModelsWencheng Ye, Anning Hu, Hongye Fang, Tianyi Wang +1
Visual perception should be iterative and task-directed, with representations refined in response to the question and the need for further processing. While Vision-Language Models (VLMs) increasingly support extended reasoning, their visual tokens are still typically acquired in a single, task-agnostic encoder pass. This creates an evidence-acquisition bottleneck: models may reason longer without refining the visual evidence needed to answer the question. We propose LoopLook, a lightweight encoder-side recurrent perception framework that equips pretrained VLMs with adaptive visual depth. LoopLook recurrently revisits a middle-to-late window of the vision encoder, injecting question representations at each loop to guide visual refinement. A learned stopping mechanism adaptively determines the recurrent depth, while a token-wise Mixture-of-Looks (MoL) combines representations from different loops, allowing each visual token to selectively integrate information acquired across recurrent steps. Across four backbones and eight benchmarks, LoopLook consistently improves performance over the baseline, yielding gains of up to 8.6% on MMStar and 8.1% on HallusionBench. Our analysis shows that LoopLook progressively refines visual attention and adaptively allocates depth according to task difficulty. It also better preserves task-relevant visual information under reduced input resolution, while increasing time to first token (TTFT) by only 5.1-6.5%. By bringing recurrence from reasoning back into perception, LoopLook transforms visual perception from a one-shot glimpse into an adaptive process of recurrent refinement.
benchmark - arxiv:2610.30822 · cs.AITESSA: A Governance-Oriented Custom GPT for SME Guidance under the EU AI Act: Design, Implementation, and EvaluationChalisa Veesommai Sillberga, Kai-Kristian Kemella, Pekka Sillberga, Katri Harjuveteläinenb +2
The increasing adoption of generative AI presents new opportunities while raising AI governance and regulatory challenges, particularly for small and medium-sized enterprises (SMEs) navigating the European Union Artificial Intelligence Act (EU AI Act). To support AI governance education, this study developed TESSA, a governance-oriented educational artifact within the OpenAI Custom GPT environment. Following a design science research approach, TESSA was evaluated against baseline ChatGPT using 100 benchmark scenarios through governance alignment assessment, human expert review, and consistency testing. Preliminary LLM-assisted assessment showed governance alignment rates of 90% for TESSA and 83% for baseline ChatGPT, while independent validation by an AI regulation law practitioner yielded 71% and 70%, respectively. Human expert review indicated that both systems provided generally strong governance guidance, while TESSA maintained high consistency across repeated executions. The findings suggest that governance-oriented grounding can support structured, practical, and stable AI-assisted compliance guidance for SMEs. The study contributes (i) a governance-oriented Custom GPT artifact for AI governance education, (ii) a human-validated regulatory update mechanism, and (iii) a multidimensional evaluation framework for governance-oriented AI artifact.
benchmarkevaluation framework - arxiv:2610.30819 · cs.ROTEAR: Task-Evidence Authority Routing for Robotic ManipulationShicheng Zhang, Hui Liu, Haoliang Li
Reliable robotic manipulation in field-deployed settings requires more than executing a capable controller. As the execution conditions change, the system must decide, online and on the robot side, which control source should hold authority. Human teleoperation adapts readily to contact-rich and semantically ambiguous situations but depends on a stable communication link. Local autonomy is executed without continuous human intervention, but may fail under semantic ambiguity, long-horizon drift, and unforeseen contingencies. These two control sources exhibit complementary reliability profiles. We formulate this complementarity as a control authority routing problem and present Task-Evidence Authority Routing for Robotic Manipulation (TEAR), an edge-deployable framework that allocates authority between human teleoperation and local autonomy. TEAR arbitrates on the robot side using network-state evidence, which governs the usability of teleoperation, and task-semantic evidence, which reflects whether autonomous execution remains consistent with the current task. A lightweight temporal arbitrator observes both forms of evidence together with robot state and current authority to time authority transfers. Across simulated and real-world trials under dynamic network conditions, TEAR achieves task success rates close to pure teleoperation while limiting teleoperation to 19.8-27.4% of control steps and improves success by 30-40% over autonomy-only execution in our settings. These results demonstrate the potential of complementary authority routing to sustain successful manipulation with substantially reduced dependence on continuous human control.
manipulationteleoperation - arxiv:2610.30817 · cs.RORobot Motion Transition with Human Motion PriorTaehyun Byun, Taemoon Jeong, Yoonbyung Chai, Woojin Wie +3
Humanoid motion-tracking policies can reproduce diverse behaviors, yet connecting independently acquired motions into a continuous sequence remains challenging. Interpolation connects boundary poses without considering whether their relative position, heading, and timing provide suitable conditions for a transition. A generative motion prior can produce richer motion, but poorly arranged boundaries may cause it to generate excessive locomotion, turning, or another unintended action. We present Spatio-Temporal Alignment with a Motion Prior (STAMP), a framework for connecting independently acquired humanoid motions through a Robot-to-Human-to-Robot process. STAMP uses human motion space to draw on rich motion priors and reuse transition knowledge across robot embodiments. Its key component is a learned alignment model that determines the boundary conditions for motion inbetweening. Given the two motions retargeted to a human rig, the alignment model ranks candidate positions, headings, and transition durations for the following motion. A human motion prior generates the transition under the selected alignment. The completed sequence is retargeted to the robot and used as a reference to train a motion-tracking policy. We evaluate the tracking success of the generated transitions on Unitree~G1 and~ROBOTIS K1. Physical robot demonstrations further show that STAMP connects independently acquired motions into long-horizon sequences. Project page: https://childtoy.github.io/STAMP-project
humanoid - arxiv:2610.30814 · cs.CVDRILL: Training World Models to Improve Policies, Not Predict PixelsZhaolu Kang, Tailong Luo, Lei Wei, Guangyuan Dong +8
World models are increasingly used to reduce environment interaction when post-training vision-language-action (VLA) policies with reinforcement learning. Yet they are typically trained to predict pixels one step ahead, even though their role here is not merely to render visually faithful videos but to produce learning signals from imagined rollouts that improve the policy. We show that this mismatch is substantial: lower pixel MSE can produce worse policy gradients, and a world model trained for policy utility can yield better downstream policies despite 17% higher pixel MSE on the next frame. We propose the Downstream-Return Imagination Learning Loop (DRILL), a bilevel framework that updates the world model to maximize the policy's return after an inner GRPO update on imagined rollouts. DRILL has two instantiations. DRILL-VWI is a closed-form surrogate, requiring no additional simulator rollouts, that reweights prediction errors by policy visitation, the group-relative GRPO advantage, and model confidence. DRILL-IMG is a full meta-gradient method that differentiates through the inner update using truncated Hessian-vector products. We show that DRILL-VWI recovers the local one-step component of the DRILL-IMG meta-gradient under score-compatibility and small-residual assumptions. Across five manipulation simulators and two VLA backbones, DRILL-IMG raises mean success rate by 13.1 points over WMPO and 7.9 points over RLVR-World, and complements RLVR pretraining rather than competing with it. More broadly, we believe video world models for VLA reinforcement learning can be productively trained by the policy gradients they produce, not only by pixel-level prediction.
vision-language-actionvlamanipulationworld modelpost-training - arxiv:2610.30810 · cs.ROAdaptive Aerial Loco-Manipulation Policy Learning via LLM-Assisted Proactive Task GenerationJingyu Ru, Xiangyuan Xie, Lixuan Jin, Jian An +7
Loco-manipulation extends the workspace of conventional manipulation by combining mobility with physical interaction, while aerial loco-manipulation further expands this capability into three-dimensional space, enabling robots to reach environments inaccessible to ground-based systems. Moving to an aerial platform, however, gives rise to a demanding control problem, since the vehicle must track trajectories accurately in three dimensions while the motion of the arm and its interaction with the environment continuously shift the center of mass, alter the inertia, and impose force and moment disturbances on the body. Conventional controllers address these effects through explicit models and estimation, whereas learning-based controllers infer dynamics from experience, yet control rewards provide no direct target for the response information retained from history, and fixed task sampling can leave persistent tracking difficulties insufficiently explored. We present ALMA, an aerial loco-manipulation framework for adaptive flight control under external arm commands. ALMA employs a contextual flight policy that adapts from recent measurements and actions, using action-conditioned prediction of future measurement representations to train its context to retain response information relevant to flight control. To expand the experience available for this adaptation, an evolutionary process uses a large language model (LLM) to analyze observed control difficulties and propose new structures of arm motion for subsequent training. In simulation, ALMA reduces position root-mean-square error (RMSE) by up to 24.7% relative to the strongest baseline, while the same policy transfers zero-shot to the aerial manipulator and reduces hover position RMSE by 79.6% during periodic arm motion relative to the real-world baseline.
manipulationmanipulatoraction-conditioned - arxiv:2610.30804 · eess.SYNumerical Solution of Large-Scale Low-Rank Coupled Generalized Sylvester EquationsUmair Zulfiqar
This paper develops projection methods for computing low-rank solutions to large-scale coupled generalized Sylvester equations. The dominant subspaces of these solutions are shown to be spanned by specific resolvent directions, motivating a Petrov--Galerkin projection framework. Conditions for the unique solvability of the projected coupled generalized Sylvester equations are also discussed. Based on these dominant subspaces, two algorithms are proposed: an extended Krylov-based projection algorithm and a rational Krylov-based projection algorithm. The former uses fixed shifts, whereas the latter uses adaptive shifts. Numerical experiments on large-scale benchmark problems demonstrate that the proposed methods achieve high accuracy and fast convergence.
benchmark - arxiv:2610.30803 · cs.LGWhat Lies Beyond Perception: A Partition Theory of the Limits of AI Evaluation and OversightAhmed M. Mahran
Replacing the judge can move a benchmark score, and models can be trained to hide capabilities. Both put the evaluator in question: what can it certify? We develop a partition theory of evaluation from one principle: judgement is bounded by perception. An agent's discrimination is the partition of its inputs into classes it cannot tell apart. Discrimination is necessary for capability but not sufficient, so bounds on it bear directly on whether evaluation can show that a model lacks an ability. We prove a one-sided bound. An examinee whose partition refines its evaluator's has the discrimination to pass every test the evaluator can grade, provided each input has an acceptable reply. We define the echo test, which accepts any reply the evaluator perceives as identical to the input, and show that it is complete for the tests the evaluator can grade. Passing therefore certifies at most a floor on discrimination, never a ceiling, and only a sharper evaluator can raise it. Used to detect concealment, echo-test failure points the wrong way: under exhaustive probing it flags every honestly weaker examinee and no concealer that presents as the evaluator's peer. Exact imitation of a weaker model is undetectable by any behavioural evaluator. We extend the bound to ensembles of monitors and to channels a model controls, explain published findings on judge disagreement and sandbagging, and derive testable predictions. A safety case that argues from behavioural evaluation alone that a model lacks an ability must assume the model's cooperation.
benchmarkevaluator - arxiv:2610.30802 · cs.LGIs spectral predictability enough? Testing Ω-matching for synthetic pretraining under distribution shiftEdric Castel C. Hao, Juan Raphael S. Seña, Neon S. De Jesus
Synthetic pretraining under distribution shift requires deciding what synthetic data to use. A natural criterion is to match the spectral predictability $Ω$ of the source to the target, but is $Ω$ sufficient? We test this hypothesis, which we call predictability-aware pretraining, across three architectures (LSTM, PatchTST, TCN) and three seeds. Under ideal synthetic-to-synthetic conditions, $Ω$-matched pretraining can outperform even an oracle baseline (+46.8% MSE reduction, LSTM), though the benefit is already architecture-dependent. On the real-world ETTm1 benchmark, however, this relationship does not hold: sine-plus-noise pretraining at matched $Ω$ yields only a marginal mean improvement +2.3% averaged over all three architectures, up to +65.5% on individual channels), whereas $Ω$ matched chaotic ODE trajectories systematically degrade performance -13.2%). The two sources agree on the direction of transfer for only 53% of configurations across all three architectures (Pearson r=-0.02), indicating that $Ω$ carries little information about which source will transfer beneficially. We further observe an architecture-dependent asymmetry in the response to the direction of the $Ω$ gap. We conclude that $Ω$ matching can help in controlled settings but is not sufficient to predict transfer across sources, and we discuss additional signal properties that a source-selection criterion may need to consider.
benchmark - arxiv:2610.30796 · cs.LGGIFT: Gradient Informed Filtering for Distillation Augmented Reasoning RLShijie Li, Yilin Gao, Tieyuan Chen, Yun Luo +2
Training large language models for complex reasoning tasks relies on two complementary signals: sparse, outcome-grounded reinforcement learning (RL) for correctness, and dense, token-level distillation for guidance. However, complementarity in supervision does not guarantee compatibility in optimization; imitation-focused distillation can conflict with the outcome-driven RL objective, and existing hybrids lack a principled resolution. We propose \textbf{GIFT} (\textbf{G}radient-\textbf{I}nformed \textbf{F}il\textbf{T}ering for Distillation-Augmented Reasoning RL), a novel method that directly addresses the allocation problem of which teacher updates to apply. GIFT performs a token-level geometric analysis, decomposing the distillation gradient into a component \textbf{parallel} to the RL gradient, which measures first-order conflict and provides a \textbf{safety} filter, and a component \textbf{orthogonal} to it, which quantifies novel directional guidance and provides a measure of \textbf{value}. This decomposition yields a principled scoring rule to rank token-level guidance. We combine this with a progressive \textbf{coverage annealing} schedule that shrinks the budget of tokens receiving distillation, ensuring that the most conflicting updates are discarded first while the most complementary ones are retained at full strength. This facilitates a smooth transition from dense, geometrically-filtered guidance to a pure, outcome-grounded objective at convergence. Extensive experiments on mathematical and logical reasoning benchmarks show that GIFT consistently outperforms strong baselines across multiple model scales, achieving relative gains of up to 9.5\% on difficult mathematical reasoning tasks. Code: https://github.com/lilifree/GIFT
benchmark - arxiv:2610.30792 · cs.LGTrace2World: Reconstructing App Worlds from Real Interaction Traces for Scalable GUI Agent TrainingZezhou Wang, Hanzhang Zhou, Panrong Tong, Quyu Kong +4
Scaling GUI reinforcement learning requires environments that preserve real-app behavior. Yet coupling behavior inference with coding can turn partial traces into runnable mocks with incorrect state transitions. We introduce \textbf{Trace2World}, which separates global behavior reconstruction from implementation. It reconciles traces into a shared, source-linked app model, then compiles an executable environment. Independent replay of source workflows localizes mismatches for bounded implementation repair. Validated worlds then support learner-adaptive task and environment augmentation within source-supported behavior. From 1,570 traces, Trace2World constructs 50 environments. On 200 workflows across ten apps, 91.1% are reproduced end to end, versus 59.8% for the strongest matched task-local baseline. Source-grounded validation retains 94.0% of simulated training gains on real devices, versus 54.7% under self-validation. Learner guidance raises benchmark success from 56.9% to 61.1%. Training in Trace2World further improves the model by 20.2 points on our benchmark, 5.4--16.2 points across four external GUI benchmarks, and 16.3 points on 100 real-device tasks. Reliable reconstruction separates what behavior source evidence supports from whether the implementation preserves it.
agentbenchmark - arxiv:2610.30791 · cs.LGGandalf: Flash Shall Not Die! Increasing Flash Storage Lifetime for KV Cache Offloading in LLM ServingMaharani A. P. Irawan, Kuntai Du, Jiayi Yao, Dongjoo Seo +3
KV caching reduces the serving cost of large language models (LLMs), but the growing KV cache footprint is increasingly exceeds GPU and host-memory capacity. Therefore, people use SSD to expand the KV cache storage capacity with high I/O bandwidth. However, in production LLM workload, such as multi-round conversation and LLM agents, the total amount of KV cache writes significantly exceeds SSD capacity, shortening SSD lifetime from years to months. To improve SSD lifetime, a natural solution is to reduce the KV cache writes by gating the KV cache writes via historical access patterns. However, that significantly lowers the KV cache hit rate (and increase the LLM serving cost) since KV caches from different attention types might have identical historical access pattern but different future access patterns. Our insight is that, to reduce KV cache writes without hurting KV cache hit rate, one need to not only consider the access pattern, but also application-level information such as the attention type of KV cache. Leveraging this observation, we propose Gandalf , the first SSD KV cache offloading system with SSD life time awareness. Gandalf prioritize the admission of full-attention KV cache into SSD over the cache generated by other types of attention, which significantly increases SSD lifetime without sacrificing cache hit rate. Our evaluation shows that Gandalf extends the SSD lifetime by 7.3X on average, while having negligible impact on cache hit rate across multiple workload traces and wide range of hardware capacity. Gandalf design can be extended to other flash-based storage, such as High Bandwidth Flash (HBF).
llm agent - arxiv:2610.30790 · cs.AIAgentic Knowledge-Model Co-Evolution via Consistency RegularizationSirui Chen, Zhining Liu, Chao Zhang, Jingrui He +1
LLM agents can evolve by learning from accumulated experience. Existing approaches extract reusable external knowledge, such as skills and memories, from experience to guide subsequent task solving, with some further internalizing this knowledge through model training. However, different ways of analyzing and transforming the same experience can yield knowledge with substantially different utility, even when the resulting guidance appears reasonable. Inspired by consistency regularization, we identify consistent utility gains across different ways of analyzing the same experience, rather than relying on any single way of analyzing it. We introduce SPLICE, a framework for consistency-regularized knowledge-model co-evolution. It constructs evolution views by perturbing how the same experience is analyzed, producing different knowledge candidates. Knowledge updates are then guided by common gain, which accounts for utility gains and their consistency across views. During model evolution, self-distillation internalizes the evolved knowledge alongside reinforcement learning, using knowledge associated with consistent utility degradation as a contrastive reference. The two processes interleave, with model updates driving further knowledge evolution. Experiments across diverse domains show that SPLICE outperforms knowledge evolution and model training baselines by up to 7.1 and 4.6 points in average accuracy, while its evolved knowledge reduces output tokens per correct solution by up to 44% and transfers across model sizes and families.
llm agentagentic - arxiv:2610.30789 · cs.AIAn Introduction to World Models for Generalist AgentsZhihui Zhu
World models have recently emerged as a promising direction toward developing generalist agents capable of understanding, predicting, and interacting with diverse environments. However, despite growing interest, the concept of a world model remains loosely defined, encompassing approaches ranging from predictive dynamics models and learned simulators to models that jointly generate future observations and actions. This introductory paper aims to clarify the concept of world models and provide a unified perspective on their role in developing generalist agents. Starting from the partially observable Markov decision process (POMDP) framework, we connect world modeling to classical control, dynamical systems, signal processing, and reinforcement learning. We introduce different approaches to world modeling, emphasizing the knowledge they capture, their learning objectives, and how they support decision-making. We further explore the relationship between world modeling and policy learning, highlighting how these capabilities can be unified through conditional prediction without requiring separate models. Rather than offering an exhaustive survey of existing methods, this paper focuses on fundamental concepts, modeling principles, and connections, with the goal of providing an accessible and unified introduction to world models for generalist agents.
world model - arxiv:2610.30782 · eess.SYAnomaly Score Preserving Data Poisoning Attacks on Data-Driven Controller DesignMotoharu Tsuboi, Hampei Sasahara, Iori Takaki, Hideaki Ishii
This paper investigates stealthy data poisoning attacks against data-driven controller synthesis and develops an attack generation method. During data collection, an attacker can sequentially manipulate state measurements, while the defender attempts to detect such manipulation using a residual-based anomaly score based on the Mahalanobis distance. Under suitable assumptions on the defender's prior knowledge, we characterize the class of time-varying affine transformations that preserve the anomaly score for arbitrary data and parameterize this class by the orthogonal group. Based on this characterization, we formulate attack design under the stealthiness condition as an optimization problem over the orthogonal group and propose a Riemannian gradient method. Numerical experiments demonstrate that the proposed attack can degrade closed-loop performance while remaining stealthy.
manipulation - arxiv:2610.30781 · cs.CVDeXplicit: Making Physical State Explicit in Dexterous Manipulation World ModelsAdam Lee, Mutian Xu, Yukang Cao, Shulin Tian +1
Before acting on an object, a robot needs to anticipate how its hand motion will change the physical world and what the resulting scene will look like. Existing approaches typically entangle action-conditioned dynamics, camera motion, and visual synthesis within a video generation model, or represent dynamics as point trajectories without explicit object structure. We introduce DeXplicit, a dexterous manipulation world model that instead predicts how the world evolves explicitly in metric 3D before rendering its visual consequences. Given an initial scene and a future hand trajectory, a lightweight 3D dynamics network predicts object-centric motion and deformation, providing structured and geometrically grounded future world states. These predictions then serve as a strong scaffold for video generation: appearance is transported through the predicted 3D motion and projected into future views, allowing the renderer to focus on completing uncertain and newly visible content rather than implicitly learning world dynamics from pixels. This explicit-3D-first decomposition also enables efficient causal generation, requiring only two denoising steps per block while conditioning on generated history. Across HOT3D, OakInk2, and PhysTwin, DeXplicit reduces moved-point error by 27-40% relative to PointWorld and improves object fidelity IoU+ by 47% on HOT3D under a shared renderer. Our design with two-step causal generation achieves a 57x speedup from our 14B bi-directional configuration in video synthesis.
manipulationdexterousworld modelaction-conditioned - arxiv:2610.30779 · cs.AIProQuant: Preserving Expert Quality over Quantity in Ultra-Low-Bit Quantization of Mixture-of-Experts ModelsJiaqi Chen, Xianglong Yan, Chengzhu Bao, Xie Chen +1
Mixture-of-Experts (MoE) large language models reduce computation through sparse activation but still incur substantial storage costs. Mixed-precision post-training quantization (PTQ) reduces these costs by quantizing different model components at different bit widths, but existing methods often struggle to preserve generation quality under tight bit budgets. We identify two key challenges: (1) nonlinear quantization errors at extremely low precision are underestimated when cross-layer propagation is ignored; (2) direct router fine-tuning may cause routing drift, and calibration gains may not generalize to downstream tasks. To address these issues, we propose ProQuant, a weight-only PTQ framework for MoE models. Prune-or-Prioritize Mixed-Precision Allocation (PoP) jointly optimizes expert pruning and bit-width assignment. It allows higher-precision experts to substitute for severely degraded ones while improving bit-budget utilization. Routing-Consistent RMSNorm Finetune (RoCo) adapts expert input representations through mathematically equivalent transformations. It compensates for quantization errors to reduce routing drift. Extensive experiments demonstrate that ProQuant outperforms existing methods across diverse generative tasks. Code is available at https://github.com/ThrowBug/ProQuant.git.
post-training - arxiv:2610.30777 · cs.LGOne Pretrained Decision Model Is Enough for Training-Free Generalist Graph Anomaly DetectionXinyu Zhao, Qingyun Sun, Ziwei Zhang, Tianchen Zhu +2
Graph anomaly detection (GAD) aims to identify irregular entities in graph-structured data, yet heterogeneous anomaly patterns and distributional differences across graphs challenge generalizable detection. Existing training-free approaches avoid task-specific optimization but primarily rely on predefined numerical scoring rules, leaving the semantic interpretation of heterogeneous anomaly evidence largely unexplored. In this paper, we propose Jev-GAD, a training-free generalist graph anomaly detection framework that bridges heterogeneous numerical evidence and pretrained decision mechanisms. To address heterogeneous anomaly patterns, we construct complementary structural, attribute, and neighborhood evidence and transform heterogeneous statistics into graph-relative anomaly percentiles. To move beyond predefined numerical scoring rules, we leverage a single frozen pretrained language model for evidence-conditioned anomaly assessment. To bridge the semantic gap between numerical evidence and pretrained models, we design a semantic Graph-to-State (G2S) transformation that organizes graph context and semantically labeled node evidence into structured decision states for non-generative option-logit readout. The resulting decision scores are integrated with four complementary statistical scores through equal-weight rank-based fusion, without GAD-specific gradient-based optimization. Extensive experiments on seven graph anomaly detection benchmarks spanning citation, social, and co-review networks demonstrate the effectiveness of Jev-GAD, achieving average AUROC and AUPRC scores of 89.18\% and 52.40\%, respectively. Further analyses validate the contributions of pretrained decision readout and semantic evidence grounding.
semantic graphbenchmark - arxiv:2610.30774 · cs.CVEvaluation of Front-End Feature Matching for Lunar Visual SLAMGuillem Casadesus Vila, Grace Gao
Lunar rovers will primarily rely on cameras for localization until navigation satellite services are deployed, and will continue to need them afterwards for relative navigation, hazard detection, and surface operations. Visual odometry (VO) and Simultaneous Localization and Mapping (SLAM) mostly estimate motion by matching features between frames. However, feature matchers are developed on urban, indoor and driving scenes, planetary studies compare them on orbital registration or within complete SLAM systems, and real lunar imagery with reference poses is scarce, so matching between rover frames on low-texture, harshly lit terrain, and its effect on VO, has not been compared. We benchmark eleven classical and learned detector-matcher combinations on nine synthetic, lunar-testbed, and terrestrial-analogue datasets, and a feature tracker, scoring pairwise pose accuracy against baseline and the trajectory error of the resulting VO. Under a common rotation-only metric, the SuperPoint pairings outperform SIFT with ratio-tested FLANN matching on both lunar testbeds and one analogue site, with smaller or unresolved differences on synthetic data. When rotation and translation direction are scored jointly, this SIFT pairing exceeds both SuperPoint pairings on four synthetic datasets, so synthetic results alone do not establish relative performance on real rover imagery. In stereo VO, the SuperPoint pairings are not consistently more accurate than SIFT, but only they keep tracking on all eight sequential datasets. Pairings with classical detectors depend on lighting while SuperPoint changes little: a low spotlight makes them more accurate than diffuse light on the same testbed views, consistent with more contrast and detected keypoints. SuperPoint with LightGlue is the fastest learned combination (16 ms per frame), and DINOv2 retrieval proposals still need geometric verification.
benchmark - arxiv:2610.30769 · cs.AILLM Multi-Agent Systems Are Vulnerable to Herding and Runaway CascadesFarnoosh Hashemi, Shiyu Ji, Joice Chen, Michael W. Macy
Large language models (LLMs) increasingly operate in multi-agent systems where they collaborate by responding to others' choices, creating endogenous feedback loops with vulnerability to social influence, herding behavior and runaway cascades. Following the design of the classic Music Lab experiment, we tested this vulnerability in large populations of LLM agents that recommended unfamiliar songs, with or without exposure to cumulative recommendation counts. Exposure led to runaway cascades that generated extreme and unpredictable collective recommendations increasingly divergent from a song's intrinsic appeal. Compared to humans, LLM populations were generally far more susceptible to herding behavior, and unlike humans, LLM cascades were less able to self-correct. This vulnerability has implications for multi-agent collaboration in recommender systems, automated evaluation, complex problem-solving, and unauthorized conspiracies.
llm agentmulti-agentagent system - arxiv:2610.30765 · cs.AIScaling Laws for Generative Reward ModelsAngel Raychev, Nathan Lile, Rafael Rafailov
We study generative reward models (GenRMs) as online preference providers for policy optimization in creative writing. Across Qwen3 reward-model sizes from 0.6B to 14B, Thinking judges trained with GRPO have higher static Gold-label accuracy than Answer-Only judges trained with SFT. This ranking often reverses when the judges train policies through online DPO: Answer-Only rewarders yield higher peak and top-five-average policy Elo in seven of eight matched settings under a fixed, learned 32B Gold evaluator. Intermediate Thinking-judge checkpoints can also outperform later, more statically accurate checkpoints as rewarders. Sweeps over rewarder size, policy size, training duration and DPO regularization show substantial benefits from reward-model capacity, including judges larger than the policy. We analyze trajectories with evaluator-specific Elo arenas, conditional match-resampling intervals and descriptive scaling fits. These results demonstrate an evaluator-rewarder gap in the studied pipelines: fixed-distribution judging accuracy does not determine downstream optimization quality. The comparison changes judge-training objective as well as response format, and the main ranking is conditional on its learned reference rather than a universal human-quality result. A limited follow-up using a different learned judge is inconclusive overall.
evaluator - arxiv:2610.30757 · physics.opticsReconfigurable Free-Space Optical InterconnectsJaekwon Lee, Carson G. Valdez, Anne R. Kroo, David A. B. Miller +1
Free-space optical interconnects can support many low-data-rate channels while avoiding the cost, energy consumption, and packaging constraints of optical waveguide- or fiber-based systems that are widely used today. We propose a free-space optical interconnect system based on reconfigurable micromirror arrays, and show that it scales favorably due to its two-dimensional scaling. We discuss the channel count scaling model, compare our demonstration with theory, and project a channel count above 10,000 to achieve 100 Tbps with slow channels (e.g., <10 Gbps). In a proof-of-concept system with a piston-motion micromirror array, we demonstrate nine channels that are simultaneously optimized. The optimization enables -21.7 dB of average worst-case crosstalk for a channel pitch equal to twice the beam radius, a smaller channel spacing than that required in flat tip-tilt mirrors. The optimization flow is further investigated for different channel spacings, simultaneous illumination, and different numbers of outputs, which generally achieve smaller worst-case crosstalk than the flat tip-tilt mirror case. Overall, the work demonstrates the advantage of piston-motion micromirrors in the proposed optical interconnects.
optical interconnect - arxiv:2610.30755 · cs.AIIs Complex Agentic Memory Always Needed? Naive RAG Is Stronger Than We ThoughtZhengye Bao, Shengtao Zhang, Haokai Zhang, Shiqi Lou +2
LLM memory is widely used in long-term conversation understanding. Agentic memory systems often use mutation, structural organization, learned retrieval strategies, or other complex operations. Is this complexity always necessary? We study Raw-RAG, a naive RAG baseline using dense retrieval and reranking over raw history to supply evidence to a reader LLM. Across LoCoMo, LongMemEval, MEME, and MemoryAgentBench, Raw-RAG ranks first among the compared memory systems in six of seven benchmark-model settings, outperforming the strongest compared agentic memory system by 0.13-5.53 points. We investigate why raw-history retrieval remains competitive. Evidence analysis shows that its higher coverage of necessary evidence can offset lower utilization. Useful evidence includes both answer values and the relations and dependency paths needed to derive them. Even complete access to transformed memory stores leaves an accuracy gap relative to raw evidence, highlighting the importance of preserving raw information. These findings challenge the necessity of complex agentic memory for the evaluated conversation tasks and establish naive RAG as an essential baseline.
memoryragagenticbenchmark - arxiv:2610.30753 · cs.CVFilmFlow: Agentic System for 3D Virtual Film CinematographyYiming Zhang, Jiarui Guo, Haojia Wei, Yifei Liu +4
Complex camera motion is central to visual storytelling, yet translating cinematographic intent into controllable, executable camera trajectories remains challenging. We introduce FilmFlow, an agentic framework that converts natural-language instructions or reference videos into explicit camera trajectories for dynamic 3D environments. To generate these trajectories, a Director agent constructs executable camera plans by composing high-level motions from concurrent atomic operations that jointly control translation, rotation, and framing. The Director is supported by three specialized agents: an Interpreter that extracts scene-grounded cinematographic intent, a Scout that selects initial viewpoints, and a Critic that evaluates rendered results and provides visual feedback for iterative plan refinement. We also introduce CinemaBench, comprising 150 Blender scenes from self-created and licensed sources, with five camera instructions per scene for 750 text-directed tasks and 60 reference videos associated with a subset of scenes. CinemaBench supports evaluation of both text-directed camera generation and video reference-directed camera generation. Across a diverse set of evaluated baselines, FilmFlow achieves the strongest benchmark performance and the highest user preference, demonstrating effective spatial reasoning, temporal coordination, and visual composition for executable 3D cinematography.
agentagenticbenchmark - arxiv:2610.30752 · cs.AIThe Delegation Divide: An Open Index of AI-Agent Delegability across Occupations and CountriesCristian Espinal Maya, Santiago Jiménez Londoño
35.2% of task time in world employment is at least partly delegable to AI agents in 2026, and only 2.5% can be handed over in full. Existing measures score exposure to generative AI, mostly for the United States. We build the Agent Delegability Index (ADI): two language models rate all 18,796 O*NET 30.2 task statements on a frozen, preregistered rubric of delegation levels and barriers; tasks are mapped to 433 ISCO-08 unit groups and weighted by ILOSTAT employment in 146 countries. Four findings. First, delegable employment rises with income, from 22.5% in low-income to 48.7% in high-income countries, a slope of 7.23 percentage points per log point of GDP per capita. This delegation divide is compositional: the physical barrier dominates the jobs poorer countries employ most. Second, at the occupation level ADI tracks exposure closely (r = 0.93 with Eloundou et al. and 0.89 with Felten et al.), above the preregistered bound of 0.80; ADI adds delegation levels and barriers on top of exposure, for the world, not a new ranking of occupations. Third, full delegation is rare and peaks for tasks that take a human about twelve minutes (11.8 minutes). Fourth, the preregistered adoption gap is not supported: across 101 countries, delegability and observed AI usage correlate at r = 0.86, and the gap between them is not wider in middle-income countries (p = 0.27). Model agreement is kappa = 0.77; a blind human rating of 300 tasks agrees with the model consensus at kappa = 0.85, and an agentic panel with web search at kappa = 0.87. We release the index, the atlas, the prompts and an MCP server at https://github.com/Cespial/delegation-divide.
agentai agentagentic - arxiv:2610.30746 · cs.LGWavenumber-Aware Deep Learning for Robust Analysis of Raman SpectraJihwan Kim, Jaeseong Heo, Yujin Kim, Sang Bin Yoon +6
Machine-learning models for spectroscopy usually treat a spectrum as a fixed-grid intensity vector, leaving the physical coordinate implicit in the array index. Such models are tied to a single sampling grid, spectral range and resolution, which limits their use when measurement conditions change. Here we represent each spectral point as a pair of intensity and wavenumber and compute Transformer positional encodings from the wavenumber itself. On Raman spectra of three bacterial species and their synthetic mixtures, this representation remained robust to noise, sparser sampling and partial spectral coverage across convolutional, residual and Transformer networks. In controlled comparisons, wavenumber-based encoding outperformed index-based and learned encodings under partial observation and retained its performance on sampling grids never seen during training, where index-based encoding degraded. These findings were reproduced on a public 30-isolate bacterial benchmark. Explicit physical coordinates offer a simple route to robust spectral analysis under variable measurement conditions.
benchmark - arxiv:2610.30742 · cs.CLUniGuard: Multi-View Guardrail Learning for Cross-Interface Safety GeneralizationViet-Thanh Pham, Thuy-Trang Vu, Junae Kim, Rollin Omari +3
Guardrail models are increasingly used as safety layers for LLMs and agentic systems, but most are trained on fixed prompt or response interfaces and become unreliable when the same safety case appears as a conversation, agent trajectory, multi-agent trace, or policy-conditioned input. We propose UniGuard, a training framework that learns interface-invariant safety representations from multiple semantically aligned views of the same safety case. We also construct the UniGuard-FT dataset by transforming 1,000 public safety cases into prompt, response, multi-turn conversation, agent-trace, and multi-agent-trace views, and train UniGuard-0.6B with classification, supervised contrastive learning, and asymmetric confidence calibration losses. Across unseen-interface benchmarks ATBench, PolicyGuardBench, and DynaBench, UniGuard-0.6B improves over its pretrained guardrail by 45.12 F1 points on average and reduces expected calibration error by 35.41 points, while remaining competitive with much larger guardrails such as Claude-Sonnet-4.5 and GPT-5.2 - despite being a 0.6B model and trained only with synthetic data.
agentmulti-agentagenticbenchmark - arxiv:2610.30738 · cs.AIAgentTriage: Replayable Security Diagnosis Across Heterogeneous Agent RuntimesJie Ying, Xutao Zhang, Ningyuan Chen, Mengce Zheng +3
When tool-use agents write files, execute code, or send data, their final responses do not explain the resulting security events, and heterogeneous runtimes expose the relevant evidence through incompatible interfaces. We present AgentTriage, an executable contract for replayable security diagnosis across heterogeneous runtimes. The contract admits conclusions whose claims cite canonical runtime events and requires abstention when mandatory evidence is missing. Runtime overlays normalize native observations, while a deterministic pipeline joins invocation and result lifecycles, constructs a typed provenance graph, selects a policy-decisive slice, and binds claims to recorded evidence. Given the same normalized trace, overlay version, and pre-run policy oracle, the diagnosis can be recomputed and its claims re-checked without language-model judgment. Across five runtimes, one contract captures all 159 applicable executions, with an audit of native and synthesized identifiers. On 120 adjudicated cases, AgentTriage achieves 0.858 grounded joint diagnosis, 0.966 required-evidence coverage, and a 0.036 unsupported-claim rate, improving joint diagnosis by 0.175 (95% CI [0.092, 0.258]) over the strongest same-schema baseline. These results quantify the precision of replayable diagnosis and the observation boundaries that require abstention. We release an open-source implementation and selected sanitized evidence at https://github.com/AsnGs/agenttriage-open-source.
agenttool-use - arxiv:2610.30732 · cs.CVSmolVLA-MoE: Sparse Mixture-of-Experts for Efficient Multi-Task Robotic ManipulationShiyang Zhang, Miao Yang, Zhengda Li, Yang Shao +3
Vision-Language-Action (VLA) models have shown promising results in multi-task robotic manipulation by leveraging visual and language understanding for action generation. However, existing lightweight VLA models employ a single dense action expert to handle all tasks, which limits their multi-task capability under constrained computational budgets. To address this issue, we propose SmolVLA-MoE, a novel framework that introduces sparse Mixture-of-Experts (MoE) into the FFN layers of the action expert in VLA models. SmolVLA-MoE replaces dense FFN layers with multiple parallel expert FFNs and a token-level router, enabling each token to adaptively select the most suitable expert for action generation. We further introduce a shared expert with a tunable weight parameter to balance the inheritance of pre-trained capabilities and task-specific expert specialization. Through sparse activation, SmolVLA-MoE activates only a single expert FFN during inference, with a total activated parameter count of approximately 500M including the frozen VLM backbone, while benefiting from approximately 800M total stored parameters. We evaluate our method on four task suites of the LIBERO benchmark. SmolVLA-MoE achieves an average success rate of 84.25%, outperforming SmolVLA (82.75%) and OpenVLA-7B (76.5%), while maintaining comparable inference efficiency. Our code is publicly available at https://github.com/dudu435/smolvla-moe.
vision-language-actionvlavla modelmanipulationopenvlalibero - arxiv:2610.30728 · cs.LGFederated Graph Retrieval-Augmented GenerationXiaobao Wang, Renda Han, Zhiliang Xia, Meng Liu +2
Graph Retrieval-Augmented Generation (GraphRAG) leverages centralized knowledge graphs to enhance large language models with structured knowledge retrieval. However, real-world knowledge is inherently distributed across multiple organizations rather than being centrally accessible. Privacy constraints further prohibit sharing entire local knowledge graphs, preventing GraphRAG from effectively retrieving and utilizing distributed knowledge. To this end, we propose Federated GraphRAG (FedGRAG), a unified framework for distributed graph retrieval and knowledge-enhanced generation over multiple clients with knowledge graphs. Specifically, we address two fundamental challenges in federated GraphRAG. First, the absence of a global knowledge graph makes unified retrieval impossible. To address this, we develop a Federated Knowledge Routing mechanism that efficiently identifies relevant clients through lightweight knowledge indices without exposing local graph structures. Second, local graph memory cannot be shared directly because it contains sensitive entities and knowledge information. We therefore propose a controlled-disclosure evidence construction method. It extracts retrieved evidence subgraphs and converts them into an evidence abstract, enabling evidence conversion and reconstruction on the server. Extensive experiments demonstrate that Federated GraphRAG consistently outperforms existing baselines in both retrieval and downstream generation tasks, establishing a new paradigm for deploying GraphRAG in federated environments.
memoryretrieval-augmentedknowledge graph - arxiv:2610.30724 · cs.CVLearning from Next-Frame Prediction: Towards Unified and Scalable Visual Representation LearningYifeng Yao, Jinghan Li, Zhihao Zhang, Yajiao Xiong +4
Video provides a natural source of supervision: what happens next serves as a learning signal for visual world modeling. Yet masked video pretraining focuses on recovering missing content from surrounding context rather than predicting the future from the past. Autoregressive methods instead predict forward one token at a time, but have yet to match joint-embedding methods in representation quality. This gap motivates us to revisit the granularity and space of future prediction. In this work, we introduce VANE, a self-supervised visual pretraining framework built on next-frame prediction in embedding space. A causal encoder summarizes past observations for a state predictor that infers the representation of the entire next frame, making temporal evolution the primary learning signal. This prediction is carried out in a fixed embedding space provided by a frozen semantic tokenizer, emphasizing scene content and structure. To retain fine-grained spatial detail alongside semantic content, a lightweight auxiliary generative decoder reconstructs future-frame patches conditioned on the predicted representations. To unify visual modeling across images and videos, we construct a local-to-global crop sequence for still images, allowing the same encoder to predict newly revealed regions from earlier crops under the same objective. Experiments show that VANE improves with model and data scaling, achieving SOTA on most anticipation, recognition, and dense perception benchmarks among unified encoders. Especially on EPIC-KITCHENS-100 action anticipation, it surpasses V-JEPA 2.1 by 7.1 points in mean-class recall@5. These results demonstrate that next-frame prediction in embedding space is effective and scalable for visual world modeling.
world modelv-jepabenchmark - arxiv:2610.30721 · cs.LGSolvable models of learning to pursue a moving targetJohn J. Vastola, Kanaka Rajan
Agents tasked with intercepting a moving target must learn how to reach it, ideally as efficiently as possible. As observers of animal pursuit behavior have noticed, different strategies are possible: on one extreme, the agent reactively moves toward the target's current location; on the other, the agent predicts the target's future location and moves directly there. Motivated by the desire to understand how such strategies might be learned, we introduce a continuous-time linear-quadratic model of open-loop target pursuit whose optimal control strategies interpolate between these possibilities. Usefully, a small number of interpretable parameters control which type of strategy is optimal, and it is possible to derive closed-form solutions for optimal strategies, policy learning dynamics, and value learning dynamics. Exploiting our model's linear structure, we find that the time scales of learning precisely correspond to the eigenvalues of certain matrices, and that relevant eigenvalue spectra generically have a gap. We show that this gap indicates that agents tend to learn how to reach their target before optimizing their movement. Our results provide a detailed mathematical characterization of pursuit behavior and its learning dynamics, which can both serve as a benchmark for empirical work, and as a foundation for more elaborate theoretical treatments of pursuit.
agentbenchmark - arxiv:2610.30718 · eess.SYIncentive Mechanisms for Welfare Maximization Under Individual Rationality in Noncooperative Dynamical SystemsYuyue Yan, Maojiao Ye, Hideaki Ishii, Tomohisa Hayakawa
Human societies are shaped by strategic interactions among self-interested decision-makers, where conflicts between individual and social interests naturally arise. This paper investigates the design of incentive mechanism for multi-agent game systems to reconcile social welfare maximization with individual benefits under economic constraints. We develop a linear-quadratic incentive framework to reshape Nash equilibria while reconciling the constraints of Pareto efficiency, individual rationality, and budget balance. The optimal target state is identified as the solution of a constrained optimization problem associated with the social welfare and aggregated payoff functions. Our theoretical analysis characterizes the conditions of incentive functions that steer self-interested agents towards the optimal target state as a Pareto-efficient Nash equilibrium for achieving the maximum attainable social welfare. We further characterize how the choice of social welfare function constrains achievable individual outcomes, revealing a structural paradox associated with priority weighting under individual rationality. Several numerical examples are presented to demonstrate the efficacy of our results.
multi-agent - arxiv:2610.30717 · cs.CVPhyEvolve: Multi-Agent State-Evolution Planning for Physics-Aware Video GenerationYang-Hao Zhou, Lixing Tan, Zhiqi Li, Guang Yang +1
Recent advances in text-to-video generation have substantially improved visual fidelity, yet existing models still struggle to understand the physical evolution implicit in user prompts, making it difficult to generate videos with continuous physical state evolution. Here, we propose PhyEvolve, a multi-agent framework that enhances natural-language prompts through structured state-evolution planning. PhyEvolve consists of three agents: a Director, a Physical Evolver, and a Composer. Given a user prompt and an optional first frame, the Director grounds the scene and disentangles physical events from observation changes. The Physical Evolver uses scene evidence to identify physical mechanisms then constructs causally linked state transitions that unfold throughout the video, whether visible or not. The Composer utilizes these transitions to enhance the original prompts. Without retraining or model modification, PhyEvolve works directly with existing generators through prompt replacement. Extensive experiments on four major benchmarks show that PhyEvolve achieves state-of-the-art performance, outperforming the cutting-edge prior agentic baseline by 7.9 percentage points on average. These results demonstrate that explicit causal state-evolution planning improves physical consistency and state continuity in generated videos through natural-language prompt enhancement.
multi-agentagenticagent frameworkbenchmark - arxiv:2610.30705 · cs.AIULME: Predictive Coupling for Mutually Reinforcing Embedding and Language ModelingYang Liu, Zhiqiang Yuan, Shu-Dong Huang, Jiancheng Lv +4
Text embedding and text generation are two of the most useful capabilities of modern large language models, yet they are typically provided by separate specialized models (an embedder and a generator) trained with incompatible objectives and deployed side by side in retrieval-augmented generation and agent systems. Recent unified embedding-generation models such as GritLM share parameters between the two capabilities, but they either switch attention regimes between modes or optimize embedding and language modeling as parallel objectives that never supervise each other. We argue that these designs are only partially unified: the two capabilities coexist on the same weights, but do not shape each other. We propose ULME, a post-training framework that unifies embedding and language modeling on a causal decoder-only LLM through predictive coupling. ULME inserts a chain of learnable slots between the query and the answer, reads a semantic embedding and a reasoning embedding from two designated slots, and trains three objectives on a shared causal backbone: a contrastive loss anchoring the semantic slot to document representations, a language-modeling loss supervising answer generation, and a stop-gradient predictive loss that aligns the reasoning slot with a pooled summary of the answer, so that generation actively shapes what the embedding encodes. Across two model families (Qwen2.5, Qwen3) and eight scales spanning 0.5B-8B parameters, a single ULME checkpoint improves MTEB by 3.3-4.1 points over the strongest open-source baseline, tightens BRIGHT and AdvBench-IR simultaneously, and remains on par with a generation-only fine-tune on standard accuracy benchmarks (MMLU, GSM8K, MMLU-Pro, HumanEval), demonstrating that genuine unification yields mutually reinforcing rather than merely coexisting capabilities.
retrieval-augmentedagentagent systempost-trainingbenchmark - arxiv:2610.30704 · cs.CVParaLin++: Accelerating Parallel Diffusion Integrator via Intrinsic Partially Linear StructureJianrong Lu, Zhiyu Zhu, Hui Liu, Junhui Hou
Training-free few-step samplers are often described as nearly quality-lossless under conventional FID/CLIP-style evaluation. It remains unclear whether that conclusion transfers to more realistic and complex generation-quality benchmarks. We first systematize the quality metrics used to evaluate training-free sampling acceleration for diffusion and flow-matching models. We find that representative training-free step-shortening methods often remain nearly lossless under FID/CLIP-class metrics, yet can suffer substantial quality degradation when evaluated with semantic-composition, dense-instruction, and long-text-rendering benchmarks such as GenEval, DPG-Bench, and LongTextBench. Parallel sampling is a notable exception: by retaining the full-step discretization and removing only the solver's temporal dependency, it can preserve near-lossless quality while accelerating sampling. Building on this observation, we propose \textit{ParaLin++}, a step-parity parallel sampler that preserves the full-step discretization while accelerating the solver iterations. Experiments across Stable Diffusion-v2, LSUN Church, FLUX.1-dev, and Ming-Flash-Omni 2.0 show that ParaLin++ preserves the full-step Euler quality profile at $2.6\times$--$4.8\times$, $1.68\times$ on FLUX, and $1.21\times$ on Ming. Project page: https://paralinplus.github.io/.
benchmark - arxiv:2610.30698 · cs.AIGovernance-State Corruption in Long-Term Memory for LLM Agents: A Cross-Model Stress TestBowen Liu, Youting Wang, Bonan Shen, Dingyan Shang
An LLM agent can retrieve a memory correctly yet use it incorrectly after its validity state changes. We study governance-state corruption: faults that alter which retained records enter model-visible context while stored content and tasks remain fixed. Across six providers, 731 benchmark-grounded bases yield 2,080 matched fault pairs. In offline control-plane checks, corruption changes the operative projection in 96.25% of pairs, while oracle-backed enforcement restores it in 100%. For downstream evaluation, a deterministically selected, non-probability LongMemEval/Memora panel resolves 298 of 300 matched triplets (238 distinct bases). Clean-minus-corrupted contrasts are 40.54 percentage points in LongMemEval binary correctness (95% base-cluster interval [30.32, 50.71]) and 30.60 points in rescaled Memora FAMA ([23.31, 37.69]); both sources also show recovery. OAKS Novel native option parity holds for all 6,714 resolved outputs. Full-cohort frozen-scorer corruption/recovery effects (-27.55/+27.51 points) remain descriptive proxies, not validated native semantic population effects. These bounded findings motivate evaluating memory control-state integrity alongside retrieval accuracy.
memoryagentllm agentbenchmark - arxiv:2610.30694 · cs.ROLive Thread Reconstruction for Autonomous ManipulationEmma Huang, Omar Hernandez, Zih-Yun Chiu, Michael Yip
Autonomous robotic surgery has drawn increased attention in recent years due to its potential to relieve surgeons of time-consuming, mentally straining, and tedious tasks, particularly during prolonged operations. One suitable task for automation is suturing, which involves tracking the thread as it deforms during needle retrieval, de-tangling, knot tightening, and other tasks. Unlike rigid-object manipulation, deformable thread manipulation has no fixed pose to track and effectively infinite degrees of freedom, making its configuration difficult to predict and requiring the thread's state to be re-estimated rather than assumed. At the same time, thread manipulations involve sequential motions that depend on past and current thread contact points, which must be accurately and continuously tracked. This paper presents an image-to-manipulation pipeline that concurrently generates live thread reconstruction while conducting safe and optimized sequential manipulations to achieve long-horizon tasks.
manipulation - arxiv:2610.30690 · cs.AIFrom Clinical Judgment to Checkable Rules: Clinician-Led Guardrails for Mental-Health ChatbotsChenhan Lyu, Kit Wislocki, Sarah Elizabeth Stoeckl, Maitreyi Sinha +1
Large language models (LLMs) deployed in mental-health contexts can be evaluated for safety failures, but the resulting judgments are often difficult to inspect or explain. Common approaches to improving safety rely on prompt engineering, LLM-as-a-Judge, and Constitutional AI. In this work, we describe a clinician-led alternative in which two clinicians define, finalize, and adjudicate the four safety checks of a rule base, which is formally verified for internal consistency and applied as a guardrail over LLM outputs. Here and in the title, clinician-led refers to this rule base, not to the fact extractor or the uncertainty default, which are ours. Every verdict of the resulting verification layer is recomputable from a named rule and sixteen facts. Against an LLM judge matched in model, revision, quantization, rule text, and decoding but without the layer's computed 988/911 fact, the largest disagreement class, the satisfied-rule class, is structured: the Llama-3.3-70B judge, on its primary prompt, demands a crisis resource from 55 of the 65 replies (84.6%) that already contain one, while the same model, asked as a fact extractor, reports a recognized resource present on 54 of the 55. The verification layer never raises that demand, by construction. We report with equal prominence where this judge is stronger: fewer false crisis alarms on the one reference-scored check. We also report the measured cost (a specification revision raised block-class outcomes on held-out boundary prompts of an over-refusal benchmark from 6.0% to 16.0%, chiefly by changing fact extraction), the clinicians' decision to keep our uncertainty default, and the analysis of the result. The contribution is not a safer verdict. It is a guardrail whose disagreements with the judge localize to named rules and facts and whose measured costs trace to named facts and the uncertainty default.
benchmark - arxiv:2610.30688 · cs.LGAn Agent Architecture with Tools and Memory for Selective Correction of Wrist PPG Heart Rate EstimatesXu Wang, Shuhan Yi, Zhen Cheng
Wrist photoplethysmography (PPG) heart rate estimates are vulnerable to motion artifacts, and corrections that reduce mean error can worsen individual windows. We propose an agent architecture with tools and memory around a fixed heart rate estimator. Pulse, motion, and limited historical evidence generate candidate estimates, whose predicted benefits and harms guide selection. An independent risk gate determines whether to retain, correct, or reject each output. We trained the architecture on PPG-DaLiA and tested it on Wrist Exercise, GalaxyPPG, and WildPPG with frozen weights. Across these four public datasets, mean absolute error (MAE), macro-averaged over ten estimators, decreased by up to 44.5%, with the largest reduction on GalaxyPPG. Beneficial and harmful corrections decreased and increased absolute error by more than 1 bpm, respectively. Their ten-estimator macro-averaged proportions ranged from 16.06% to 36.64% and from 1.68% to 6.77%, respectively, across the four datasets. At 60% and 80% target coverage, numerical correction on the same accepted windows contributed more to total error reduction than window selection. We implemented real-time wearable acquisition, Bluetooth reception, and smartphone deployment of a selected estimator and the framework. In backend analysis of four prototype recordings from one participant, ten-estimator macro-averaged MAE decreased from 9.09 to 4.45 bpm without additional rejection. Under the evaluated conditions, the architecture improved average accuracy while quantifying correction benefits and harms, supporting wearable heart rate analysis that considers error, correction risk, and coverage.
memoryagent - arxiv:2610.30680 · cs.CVZero-Shot Domain Generalization via Domain Prompt LearningShiming Chen, Shuhuang Chen, Salman Khan, Fahad Shahbaz Khan
Domain generalization (DG) aims to improve the generalization of the model to any unseen domains. Targeting this goal, existing DG methods mainly focus on learning domain-invariant features or prompts from the given source domains. However, these methods rely on source-domain data, and collecting and annotating large-scale single- or multi-source datasets is often infeasible, limiting the practical applicability of DG. To tackle this challenge, we propose {\underline{Do}}main \underline{P}rompt Learning (DoP), a method built on a large vision-language model (e.g., CLIP) for zero-shot domain generalization, which recognizes unseen classes in unseen target domains without access to source-domain data. DoP essentially learns the target domain knowledge via prompt both in text and visual modalities to overcome the domain shift in text-visual modalities under the test-time adaptation formulation. Specifically, DoP first takes text domain learning (TDL) to simulate various distribution shifts by learning diverse style tokens implicitly in the text space with inter-style diversity loss and intra-style consistency loss. Subsequently, DoP employs visual domain learning (VDL) to further refine the target-domain-relevant style token generated by TDL by making use of the test visual data with self-supervised learning. Extensive experiments on popular DG benchmark datasets (e.g., PACS, VLCS, OfficeHome, and DomainNet) and ImageNet show that our DoP consistently outperforms the state-of-the-art DG approaches. Codes are available at: https://github.com/shiming-chen/DoP.
benchmark - arxiv:2610.30679 · cs.CVAgentDreamer: Agentic 3D Motion Reasoning for Text-Controllable Physics SimulationMijeong Kim, Suho Park, Junha Park, Bohyung Han
We propose AgentDreamer, an agent-guided optimization framework that aligns text instructions with simulated motion through explicit 3D motion planning. Given a text instruction and scene geometry, an off-the-shelf agent translates the instruction into a structured semantic motion prior, specifying what should move, in which direction, and in what temporal order. The agent represents this prior using parametric functions over a series of temporal chunks, yielding dense 3D trajectories. We then optimize the parameters of a differentiable MPM simulator such that its rollout best matches the prior while remaining constrained by the scene dynamics. This allows the agent to provide semantic guidance from language, while the simulator grounds it in physically feasible motion. We further introduce Motion-IG, an information-theoretic metric for text-motion alignment based on masked text recovery. By comparing the likelihood of recovering masked motion-related content with and without the simulated motion, Motion-IG measures the information contributed by the motion beyond that available from the remaining text and scene context. Experiments show that AgentDreamer improves text alignment over video-based guidance, producing simulations that more closely follow the requested physical behavior.
agentagentic - arxiv:2610.30676 · cs.AISpectral Smoothing and Spiking Perturbations Reconfigure Capability Profiles and Selective Task Performance in Large Language ModelsQince Luan, Jianhui Xu, Yang Xu
Singular-value spectra encode important structure in transformer feed-forward projections, yet their causal role in downstream behavior remains underexplored. We study post-hoc spectral smoothing and spiking interventions on the up_proj and down_proj matrices. Positive perturbation stretches deviations from the geometric mean, making large singular values larger and small ones smaller (spiking), and vice versa. The intervention preserves singular vectors and the geometric mean of the singular values. Across five open Llama and Qwen instruction models, we evaluate the perturbed systems on capability-labeled subtasks, standard and chain-of-thought reasoning benchmarks, judge-scored truthfulness, hallucination, and safety tasks, and TRAIT personality probes. To interpret score shifts, we fit responses in a three-dimensional capability basis consisting of factual knowledge, language understanding, and deductive reasoning. The induced changes suggest structured rather than purely random sensitivity: large positive perturbations more reliably suppress language and deductive responses, moderate negative perturbations can sometimes improve factual or task-local performance on the evaluated grids, and sensitivity varies strongly by model family, benchmark module, and underlying layerwise spectral inequality. These results position spectral perturbation as a lightweight intervention for probing how capabilities are organized and traded off in large language models.
benchmark - arxiv:2610.30671 · cs.ROCo^2Gen: Collaboratively Generating Task Plans for Decentralized Multi-Robot and Human CollaborationYi Loo, Siddhant Shrivastava, Minh Long Pham, Malika Meghjani
Visual Language Action (VLA) models are designed to be generalist robot models that can be implemented across embodiments, tasks and environments. However, their ability to collaborate with other robots and humans remains underexplored. In this work, we propose a framework to automate the generation of multi-robot collaborative data for decentralized VLA training without requiring human teleoperation. Our framework first utilizes a decentralized collaborative planning approach to generate multi-robot collaborative robot plans in the form of an executable function. We then utilize the collaborative plan to produce collaborative demonstration data to finetune VLA models in a decentralized manner. We show that our approach produces demonstration data that when used to finetune VLA models on decentralized collaboration tasks matches the task performance of VLA models finetuned on human-teleoperated collaborative data.
vlavla modelteleoperation - arxiv:2610.30668 · cs.LGRiboBridge: protein-conditioned de novo RNA binder structure designZhou Zhang, Hanqun Cao, Minghao Sun, Odin Zhang +4
Designing RNA molecules that bind to target protein structures requires generating both an RNA conformation and its bound complex geometry, without assuming a predefined binding site. Existing approaches generally model these coupled processes within a unified representation, entangling RNA conformation generation with protein-relative placement. In this work, we introduce RiboBridge, a structure-first framework that separates RNA conformational generation from protein--RNA assembly through cross-frame interaction geometry. RiboBridge first proposes multiple candidate RNA-binding interfaces on the protein surface, generates canonical RNA conformations within a pretrained RNA structural latent space (RiboSphere), and then predicts protein--RNA interaction constraints to bridge the independently defined RNA and protein coordinate frames. The resulting interaction geometry is converted into a geometrically consistent protein-relative pose through rigid placement. We evaluate RiboBridge on a temporally split, sequence-filtered, and redundancy-controlled benchmark of 124 protein--RNA complexes. Under post-hoc interface-correspondence evaluation, where native information is used only after generation to select among already generated predicted-pocket candidates, RiboBridge achieves the highest fNAT and DockQ and the lowest interface and ligand RMSDs among compared methods. Component ablations and stage-wise diagnostics further characterize the contributions of RNA conformational generation, interaction prediction, and rigid placement, identifying the main sources of complex-recovery error.
benchmark - arxiv:2610.30667 · cs.LGFast Iterative Combinatorial Auctions via Deep Reinforcement LearningLuis F. Paul Tosta, Quinn Arnold, Gianluca Brero
Iterative combinatorial auctions are used in spectrum allocation, transportation, and industrial procurement, but rounds are costly. We introduce FastCombo, a deep reinforcement learning approach to sequential price discovery. Bayesian pricing has sharply reduced auction rounds even against subgradient benchmarks tuned with hindsight for each instance, but relies on a one-step objective and repeated online inference and optimization. We prove that greedily maximizing immediate clearing probability can require arbitrarily more rounds in expectation than an informative policy, even with the true valuation distribution known. FastCombo learns price policies offline, assigning credit to queries for the future decisions they enable and producing each online price update with a single neural-network forward pass. We formulate the objective of minimizing expected rounds to clearing in single-minded auctions as a partially observable Markov decision process and train a recurrent bidder-item graph policy over complete auction trajectories. In a one-item experiment, all five trained policies learn an informative opening, reducing expected rounds from 1.85 to 1.65 relative to the tested Bayesian one-step rule. Across four domains from the Combinatorial Auction Test Suite, FastCombo achieves competitive clearing while reducing measured online price-selection time per auction by a factor of roughly 300-900 relative to the standard Bayesian baseline. These results demonstrate the value of learning sequential price policies for both information acquisition and online computational efficiency.
benchmark - arxiv:2610.30664 · cs.CLCompiling Autoregressive LLM Control into Policy-Exact Parallel Inference for Hierarchical RAGZehao Lin, Qianfeng Shen, Xixuan Hao, Shaobo Cui +7
Large language models increasingly act as controllers in structured decision pipelines, yet their decisions are still executed by autoregressive text decoding. Hierarchical RAG is one such setting: at each frontier of a document tree, an LLM commits to $b$ of $k$ sections under an input-dependent action grammar, executed as either failure-prone free-form JSON or reliable but strictly serial grammar-constrained generation. We show that this serialization is unnecessary for bounded-horizon control: the reachable future forms a shallow action-prefix tree, which we compile --- one root prefill plus one tree-causal probe evaluate every prefix-conditioned action distribution with serial decoding's tokens, positions, and causal ancestry. A prefix-state lemma makes compiled control policy-exact in exact arithmetic: the same policy with a two-forward critical path and zero output tokens. One further probe pass certifies the exact order-marginalized MAP set. Across four document-disjoint splits and two scales on the primary MetaX device, compiled control matches the serial policy on $84$--$89\%$ of questions --- every instrumented divergence a floating-point near-tie --- and cuts control latency by $35$--$47\%$ versus token-wise serial decoding and $6$--$7\%$ versus an idealized chunked engine; an H800 repetition preserves the ordering.
rag - arxiv:2610.30663 · cs.AIGroup Relative Policy Optimisation Improves Multilingual Speech Recognition in Low-Resource LanguagesTim Wenjie Hu, Evelyn Nafula Ouma, Benjamin Akera, John A. Quinn
Autoregressive encoder-decoder speech recognition models trained with teacher forcing can suffer from exposure bias, limiting their ability to recover from their own prediction errors at inference time. Recent work has shown that reinforcement learning with Group Relative Policy Optimisation (GRPO) improves speech recognition in major languages with LLM-based models. We extend this line of work to smaller, non-LLM encoder-decoder ASR models, using Whisper-large-v3 as a case study across 51 African languages. The benefits of this approach are most pronounced when training data is scarce or when input audio is noisy or degraded. Our resulting model, SunflowerASR, achieves the best published macro-averaged WER on the AfriVox-v2 (0.227) and SimbaBench (0.372) benchmarks, and is publicly available.
benchmark - arxiv:2610.30660 · cs.AIA Topology-Guided GeoAI Agentic Framework for Urban Flood ScreeningYutian Tao, Haoan Feng, Xin Xu, Leila De Floriani
Open-world geographic artificial intelligence (GeoAI) agents can transform public geospatial data into accessible decision support for real-world challenges, but must balance autonomous reasoning with reliable spatial analysis and human oversight. We present a topology-guided GeoAI agentic framework for urban flood screening that takes a place name, drawn region, or uploaded boundary and returns ranked candidate basins with an interactive map and report. A deterministic controller retrieves public elevation, rainfall, soil, land-cover, exposure, and infrastructure layers, extracts terrain depressions with discrete Morse theory, and ranks basins with a relative flood-priority score. The language model acts only at recorded decision points: selecting a persistence threshold from a precomputed candidate table, drafting recommendations from computed evidence, and reviewing each recommendation against the same evidence. Three human-in-the-loop checkpoints let users revise the threshold, factor weights, or recommendation. In a College Park, Maryland case study, the selected threshold reduced 278 raw terrain minima to 70 basins, with 77% of these basins' lowest points also identified as sinks by a separately computed eight-direction (D8) flow-routing model. Across four further study areas, agreement ranged from 61% to 83%. By combining open data, explicit spatial reasoning, traceable evidence, and human oversight, the framework enables transparent first-pass flood screening while keeping consequential recommendations inspectable and user-controlled.
ai agentagentichuman-in-the-loop - arxiv:2610.30657 · cs.AIAnticipating Social Consequences with Distributional World ModelsViet-Thanh Pham, Thuy-Trang Vu, Zhuang Li, Lizhen Qu +1
Social world models predict how interactions develop after an action, helping agents negotiate, cooperate and mediate. In these situations, one action can lead to several plausible futures, hence predicting only one future can overlook consequential alternatives, while exploring them through repeated simulations increases inference cost. We propose DSWM, a Distributional Social World Model that predicts alternative responses, their probabilities and immediate and final outcome values in one output per candidate action. We train a 4B model on repeated simulations using supervised fine-tuning and GRPO. At inference time, the actor uses these possible consequences to choose an action. Across three seen and two unseen environments, with interaction backbones absent from training collection, DSWM achieves the best average performance on 13 of 15 metrics. Ablations show that retaining multiple futures and their predicted probabilities improves outcomes in all five environments.
world model - arxiv:2610.30643 · cs.CVAsking Less, Adapting Better: Feedback-Efficient Test-Time Adaptation for Vision-and-Language NavigationZixu Wang, Huaquan Zhang, Shoukai Xu, Jinwu Hu +2
Vision-and-Language Navigation (VLN) requires agents to follow instructions and navigate to target locations. However, current VLN agents often suffer severe performance degradation in unseen environments due to distribution shifts. Test-Time Adaptation (TTA) offers a promising paradigm to bridge this gap, yet existing methods that rely on unsupervised entropy minimization can propagate errors through the trajectory, causing cascading failures in navigation. Recent methods mitigate this by incorporating human feedback, but they apply trajectory-level supervision uniformly across all steps without localizing the actions is truly informative, leading to inefficient feedback utilization and mismatched optimization granularity. To address these limitations, we propose FETA, a Feedback-Efficient Test-Time Adaptation method that improves both feedback efficiency and optimization reliability. To reduce unnecessary human interventions, we introduce a hierarchical sample selection strategy to localize informative samples. Specifically, we first filter trajectories whose normalized policy entropy exhibits large fluctuations. Within each selected trajectory, we further select the critical action via an entropy-gradient saliency criterion and query a single binary feedback on its correctness. A discrepancy-aware entropy optimization objective then updates the policy only at the queried step, with adaptive scaling that assigns larger relative weights to samples with greater prediction-feedback discrepancy. Extensive experiments across multiple benchmarks demonstrate that FETA achieves state-of-the-art performance with fewer human interventions, while remaining robust under noisy feedback and continual adaptation settings.
benchmark - arxiv:2610.30639 · cs.CVCODI: Improving Self-Supervised Vision Transformers with Cross DistillationSiran Dai, Qianqian Xu, Peisong Wen, Yang Liu +2
Self-supervised Vision Transformers learn strong image-level representations, but their patch-level representations can be unreliable for dense prediction. We study this mismatch and find that the class token concentrates global semantics on a small number of representative patches, while dense prediction errors often occur on low-attention regions. We provide a local gradient analysis showing that, along the [CLS] attention readout, image-level self-distillation gradients are weighted by [CLS]-to-patch attention mass. To provide masked patches with global semantics without changing the architecture, we propose Cross Distillation (CODI), which mixes a small amount of the teacher class-token distribution into each teacher patch-token target. CODI gives every masked patch an attention-independent semantic correction while preserving the standard self-distillation architecture. Experiments on dense and image-level benchmarks show that CODI improves both dense and image-level performance under aligned pretraining settings. Code is available at https://github.com/EldercatSAM/CODI.
benchmark - arxiv:2610.30635 · cs.CLWhat the Specification Did Not Say: Specification-Oracle Gaps in Test-Graded Coding-Agent BenchmarksFei Ding
Coding-agent benchmarks often describe a task in natural language but grade it with tests taken from the upstream project. Passing those tests can require information the description never states: a constant's name, a module path, an exact message. We call this a specification-oracle gap: where it exists, a benchmark also measures whether an agent reproduces upstream implementation details. We formalise the gap, including the inconsistencies that chains of dependent tasks add, and build a static audit for five languages that classifies every name-, path- and literal-level dependency of every scored test (held-out precision 89.2%, recall 92.9%). On SWE-Milestone (98 graded milestones, 35 published runs), 25 milestones contain a test that depends on an unstated, non-inferable name or path. Build units with such a dependency fail in 80.6% of runs, against 48.4% for comparable units without one, and withholding a single hidden name from the reference stops its unit's tests from compiling in 126 of 128 cases. The leaderboard's ranking is robust to removing the affected tests, but they shift scores within repositories and favour agents that reproduce upstream names. Across four further benchmarks, prevalence appears to follow construction: a third or more of the tasks are affected where specifications are release notes or LLM summaries, almost none where tasks are screened or written for their tests; how much a gap costs depends on how the harness grades. We release the audit, corrected filter lists and specification addenda.
agentagent benchmarkbenchmarkleaderboard - arxiv:2610.30628 · cs.LGFuture of Time Series Anomaly Detection: From Historical Review to Unified OutlookZhijie Zhong, Zhiwen Yu, Pengyu Li, Liujian Zhang +5
Time series anomaly detection (TSAD) is a foundational problem in data mining, aiming to identify observations that deviate significantly from nominal temporal patterns. Driven by escalating data volumes and application complexity, recent research has yielded diverse architectures, including diffusion models, state space models, and large language models (LLMs). However, existing surveys provide limited coverage of these emerging designs and lack a unified treatment of multivariate anomalies, cross-domain unified modeling, metric reliability, and contemporary evaluation challenges. To bridge these gaps, this survey delivers three core contributions: (1) We establish a refined anomaly categorization and a structured taxonomy of paradigms reflecting modern advances, comparing detection mechanisms and charting developmental trajectories across paradigms. (2) We provide an extensive analysis of evaluation methodologies and conduct comprehensive experiments to assess metric fidelity, complemented by an open-source benchmark repository encompassing algorithms, datasets, and evaluation tools. (3) We outline key open challenges, offering forward-looking perspectives on dynamic thresholding, model interpretability, zero-shot transfer, multi-modal alignment, and foundational models for time series. By addressing existing theoretical and empirical limitations, this survey establishes a coherent framework to guide future TSAD research and practical deployments.
benchmark - arxiv:2610.30611 · cs.AIBetter Reasoning, Not Just Bigger Models: A Structure-Grounded Framework for Efficient and Reliable InferenceAmirhossein Aghaei, Liangyan Li, Shiyi Yang, Shaohua Wang +1
Scaling large language models has substantially improved linguistic fluency, yet reliable autonomous agents require more than fluent generation. In multi-turn inference, current agentic systems often produce semantically plausible but structurally ungrounded reasoning, especially under incomplete observations. We propose Structure-Grounded Uncertainty Reduction (SGUR), a training-free framework that complements token-level generation with active uncertainty reduction over structured hypothesis spaces. SGUR represents incomplete observations as a query graph and aligns them with candidate hypothesis graphs through a prior-weighted partial graph matching mechanism, where database-derived empirical priors assign greater importance to decision-relevant relations rather than treating all graph edges uniformly. The resulting inference state is iteratively refined by an active information-gain strategy that selects the next most informative node to query, systematically pruning the candidate space. We instantiate SGUR on a challenging PrimeKG-derived sequential medical diagnosis benchmark, demonstrating that structural grounding can partially substitute for model scale: starting from only two observed symptoms, SGUR achieves 92.5% Hit@3, outperforming the strongest Graph-RAG baseline by 9.4 percentage points while using approximately 5x fewer tokens. Notably, SGUR enables smaller backbone models to match or exceed the performance of models up to 10x larger, suggesting that structural grounding provides a more robust and efficient paradigm for scaling agent performance than relying exclusively on increasing model size.
agentautonomous agentagenticbenchmark - arxiv:2610.30609 · cs.CVKinoUAV: A Benchmark for Safe and Stable UAV Vision-Language Navigation with Kinodynamic ModelingChenyao Yang, Peijin Wang, Yingchao Feng, Wenhui Diao
Unmanned aerial vehicles (UAVs) have recently emerged as important remote sensing platforms, and UAV navigation has shown great potential in remote sensing applications. However, existing methods do not explicitly incorporate kinodynamic modeling or safety distances into trajectory generation, resulting in high absolute yaw rates, frequent collisions, and low navigation success rates, which limit safety and motion stability. To address this challenge, we introduce KinoUAV, a kinodynamic benchmark comprising a dataset and KinoUAV-Agent. First, we develop a hierarchical kinodynamic trajectory generation pipeline that drives UAV through acceleration actions. It uses waypoint chains from global planning and generates trajectories segment by segment under velocity, acceleration, and jerk constraints, reducing oscillations and improving trajectory stability. Furthermore, we introduce directional free-space distance (DFD) labels into the constructed dataset to describe the available clearance for safe motion decisions. The proposed KinoUAV-Agent predicts DFDs and uses them to generate safe actions, decoupling perception from action decision-making. This design enables autonomous obstacle avoidance and improves navigation safety. The benchmark incorporates the proposed Cruise-Seek task, in which the UAV first enters the target region and then searches for the target, reducing the difficulty of target search and improving navigation success. Experiments on KinoUAV show that KinoUAV-Agent outperforms existing methods, achieving the highest success rate while maintaining lower absolute yaw rates and collision rates. These results demonstrate the effectiveness of kinodynamic modeling for safe and stable UAV navigation.
benchmark - arxiv:2610.30607 · cs.ROContinuous-Time Motion Planning and Steering with Global ConvergenceMohsen Amiri, Sima Didari, Davoud Ataee Tarzanagh, Hankyu Moon +1
Motion planning is central to robotics, and it is increasingly used to correct the motions proposed by learned planners and policies. In both uses, constraints must hold at every instant of the motion, not only at sampled times. To our knowledge, no algorithm provably converges to a globally optimal motion once a smoothness cost is combined with such constraints. We cast planning and steering as a single continuous-time program: robots avoid one another and static or moving obstacles, and a learned proposal is steered by keeping the motion close to it. The program is elementary to state, yet nonconvex. We propose GlobMP (Global Motion Planning), a gradient projection method over piecewise-polynomial curves, on which each constraint is one polynomial per piece that must stay nonnegative over the whole motion; it alternates cost reduction with feasibility restoration. Under mild assumptions, we show that, from a fine grid start, its cost approaches the global optimum as the number of polynomial pieces grows, with every constraint met to any chosen tolerance, or exactly on stable instances. We demonstrate its superior performance over strong baselines in real and synthetic experiments, including teams of up to $200$ robots, multi-robot planning on the Robotarium testbed and an extended benchmark, maze navigation, and steering tactile-reactive policies on a bimanual humanoid.
humanoidtactilebenchmark - arxiv:2610.30606 · cs.ROCMAR-search: Commonsense and Memory Augmented Reasoning for Object Search in Dynamic Interactive EnvironmentsKaiyao Liao, Qingfeng Li, Xinlei Zhang, Chen Chen +2
Dynamic interactive object search in large-scale human environments presents substantial challenges for existing methods. Current scene representations like 3D Scene Graphs (3DSG) only provide coarse-grained spatial segmentation and cannot identify functional areas such as storage or leisure areas. Without functional area understanding, existing methods are constrained to exhaustive sequential exploration at large scales, resulting in inefficient search behaviors -- particularly in open-layout environments with numerous interactive objects such as drawers and cabinets. Moreover, these methods lack adaptability to environmental dynamics such as object relocations. To address these limitations, this paper proposes CMAR-search, a novel framework built upon Commonsense and Memory Augmented Reasoning (CMAR). Our approach leverages commonsense about area functionalities and aggregates environmental memory to construct a Functional 3D Scene Graph (F3DSG), which organizes the environment into functional areas with their associated containers. Through this structured representation, CMAR enables hierarchical action planning at both macro-area and micro-container levels, empowering the system to efficiently identify and inspect semantically relevant areas for effective object search. Notably, CMAR continuously integrates real-time perception, accumulated memory, and commonsense to dynamically relocalize objects in changing environments. Extensive experiments in simulation and real-world settings demonstrate that CMAR-search significantly surpasses state-of-the-art baselines in both success rate and search efficiency for object search in dynamic interactive environments.
memoryscene graph - arxiv:2610.30605 · cs.ROPATH: Preference-Aware Traversability Estimation for Robotic Teleoperation in Off-Road NavigationXianghao Meng, Wenshuo Wang, Cheng Shao, Lewei Feng +3
Traversability estimation is critical for ground vehicle teleoperation, yet existing methods often neglect the human operator's preferences for speed and risk tolerance. To address this gap, we propose PATH-Net, a Preference-Aware Traversability estimation Hierarchical Network that adaptively generates preference-conditioned traversability cost maps and speed profiles for off-road teleoperation. Our framework integrates visual and geometric bird's-eye-view features with short-term historical vehicle states through a multi-head shared-backbone architecture. Our PATH-Net infers traversability cost maps while optimizing speed recommendations by using maximum entropy inverse reinforcement learning and sparse speed supervision. Our method incorporates operator preference via an asymmetric feature aggregation mechanism, splitting inference into a cost branch that prioritizes high-risk traversability regions and a speed branch that outputs preference-adjusted recommended speeds. Experiments with 15 operators across 225 teleoperated trajectories show alignment with user-specific navigation strategies, achieving modified Hausdorff distance of 0.17 m (aggressive) and 0.15 m (conservative). In the questionnaire assessment, participants reported lower NASA-TLX workload scores with PATH-Net recommendations than without recommendations, decreasing from 34.7 to 22.2. These results support preference-conditioned trajectory and speed recommendation for off-road teleoperation.
teleoperation - arxiv:2610.30593 · cs.CLTT-SAX: A Two-Track Time Series Symbolization for Natural Language Shape QueryingZhao Tan, Xianglong Bao, Zhipeng Xi, Ming Jin
Natural language querying for time series databases aims to let users find what they need in long histories of data by asking in plain language. Shape queries ask how a series moves. They are the main challenge of this task. Recent work makes them feasible by using symbolization as a bridge between the query and the long history. However, existing symbolizations were designed for compression, similarity indexing, or classification. What a symbolization should provide for natural language querying has not been systematically studied. We therefore examine the definitions of six classical symbolizations against this task and identify three limitations. (1) Direction is missing from the string, (2) direction is measured at too small a scale, and (3) the letters are arbitrary. Three design principles follow. We follow these principles and make a minimal redesign of SAX called TT-SAX (two-track SAX). The level letters stay. Each segment gains a trend letter. It is measured across neighboring segments rather than inside one, and graded to match the adjectives that queries use. We conduct extensive experiments on the public NLQTSBench benchmark, comparing TT-SAX with the six classical symbolizations across six LLMs. TT-SAX outperforms all six on shape queries, and ablations trace its gains back to the three limitations one by one. This paper is the first systematic study of time series symbolization with natural language querying as the design goal. It provides a foundation for future research on this task. Our code is available at https://github.com/Atlamtiz/tt-sax.
benchmark - arxiv:2610.30592 · cs.CVVisual Grounding First, Multimodal In-context Learning FollowsMinhyuk Seo, Minjae Lee, Chaeeun Lee, Wei Lin +3
LLMs have shown strong in-context learning (ICL) capability, but extending it to Multimodal LLMs (MLLMs) remains challenging. Prior multimodal ICL methods often rely on ICL-specific datasets for additional training or task vector extraction, which can improve performance on the same ICL benchmark used for adaptation. However, they do not generalize well to other benchmarks and often induce forgetting in previously well-solved tasks. For this reason, we identify 'visual grounding' as the key bottleneck in multimodal ICL; MLLMs often fail to attend to task-relevant visual evidence in demonstrations, instead over-relying on textual cues and producing hallucinated outputs. To address this, we propose ProCoRe to enhance visual grounding through contrastive reinforcement learning on multimodal contrastive data for improving multimodal ICL. ProCoRe generates captions for contrastive image pairs, decomposes them into propositions, and optimizes contrastive rewards that encourage alignment with corresponding images while discouraging mismatched ones. Despite never seeing ICL-formatted training examples, ProCoRe improves average multimodal ICL classification accuracy by 5.6% and captioning ROUGE-L by 13.8% over the off-the-shelf Qwen-VL-3-8B model, outperforming ICL-data-dependent state of the arts across benchmarks. Finally, we introduce PICL, a personalized multimodal ICL captioning benchmark that tests whether MLLMs genuinely understand visual demonstrations rather than relying on pre-trained semantic priors.
benchmark - arxiv:2610.30589 · cs.ROOASIS: Online Anchor-centric Scene Inference System for Open-Vocabulary Retrieval and Navigation in Dynamic EnvironmentsKaiyao Liao, Qingfeng Li, Chen Chen, Xiaoze Wu +2
Reliable robotic mapping in dynamic real-world environments is essential for embodied intelligence and long-term autonomy. However, existing methods often involve a trade-off: lightweight online mapping is efficient but captures limited semantic context, while dense mapping provides richer semantics at higher computational cost and is more vulnerable to distractors. They also struggle to relocalize objects after scene changes. To address these challenges, we propose the Online Anchor-centric Scene Inference System (OASIS), a unified system for online mapping, complex open-vocabulary retrieval, and navigation in dynamic environments. Motivated by the observation that indoor scenes are organized around stable semantic anchors, OASIS represents the environment using a sparse set of anchor nodes, each maintaining a unified multimodal representation that integrates semantic descriptions, visual features, and geometric states for reasoning, verification, and navigation. Its compact anchor-centric memory supports on-demand evidence extraction and hierarchical retrieval, reducing redundant storage while improving retrieval reliability. To handle scene dynamics, OASIS further reuses transient observations collected during navigation to relocalize moved objects. Experiments in both simulation and real-world environments, including standard retrieval benchmarks and navigation tasks under both static and dynamic settings, demonstrate that our system can perform complex open-vocabulary retrieval and navigation, while also enabling relocation of displaced objects after scene changes.
embodiedmemorybenchmark - arxiv:2610.30587 · cs.ROKnow When to Escalate: Critic-Guided Adaptive Execution for Cloud-Local Embodied AgentsWei Zhang, Lirong Xiang, Charles Fleming
Recent advances in foundation-model agents have enabled increasingly capable embodied systems for long-horizon planning and interaction. However, embodied agents require fast, closed-loop decisions as they continuously perceive and act on a changing environment. Deploying strong cloud agents leaves the platform idle for seconds between consecutive actions, while lightweight local models respond quickly but are more vulnerable to errors. Existing hierarchical architectures typically assign models fixed roles, whereas conventional model routers make stateless per-request decisions. We present CARE, a critic-guided adaptive execution framework that dynamically determines when to escalate control from a local executor to a stronger cloud model. A lightweight critic model monitors execution reliability and triggers escalation when local execution becomes unreliable; once escalated, control persists until the agent recovers, avoiding step-wise switching. To support such control transfer, a shared handoff memory maintains execution context and task-level progress across executors, enabling the model to intervene without restarting from scratch. Across diverse embodied benchmarks, CARE improves task success by 6.3-18.3 percentage points over the strongest evaluated collaboration baseline. CARE further approaches or exceeds Cloud-only success while reducing median time-between-action latency by 28.8-82.6%.
embodiedmemoryagentembodied agentbenchmark - arxiv:2610.30579 · physics.opticsEfficient third-harmonic generation in high-Q tantalum pentoxide microcavitiesXinke Xing, Yixuan Yang, Xinzhi Zheng, Renhong Gao +3
Nonlinear harmonic generation in microcavities offers a compelling route toward energy efficient on-chip short-wavelength coherent light sources, enabled by tight optical confinement within compact mode volumes and flexible dispersion engineering. Tantalum pentoxide (Ta2O5) photonic platforms are particularly promising for realizing such nonlinear processes: their broad transparency window (280-8000 nm) supports broadband operation, their large bandgap suppresses two-photon absorption, and their thick-film deposition capability enables dispersion management--features that can collectively enhance the performance of nonlinear light sources. Here, we demonstrate, to the best of our knowledge, the first efficient third-harmonic generation (THG) in high-Q Ta2O5 microcavities. These microcavities feature smooth surfaces and high loaded Q factors up to 2.67X10^6, thereby facilitating nonlinear interactions. Through dispersion engineering, efficient THG is realized in both suspended microdisks and monolithic microrings under telecom-band continuous-wave laser pumping, as confirmed by optical spectra, visible-light microscopy, and the near-cubic pump-power dependence of the conversion efficiency. This work lays a foundation for chip-scale f-3f self-referencing of frequency combs and next-generation integrated nonlinear photonics.
microring - arxiv:2610.30577 · cs.LGEvaluating Type-1 Diabetes Digital Twins by Decision QualityVictor Li, Owen Tucker, Michael S. Hughes, Temiloluwa Prioleau +1
A digital twin (DT) for type 1 diabetes (T1D) can predict how an individual's blood glucose will respond to different insulin treatments without actually trying them. For each treatment, downstream statistics of predicted glucose trajectories, such as glycemic risk, can be computed to guide clinical decisions. However, the quality of DTs is usually measured by the accuracy of their predicted glucose trajectories, which may not reflect the quality of the downstream treatment decisions they inform. In this work, we directly evaluate DTs by this decision quality on downstream treatment ranking, i.e., whether a DT predicts glucose trajectories that can be used to rank candidate treatments as the real patient would. We present an evaluation protocol over 30 virtual patients from the FDA-accepted UVA/Padova model, each simulated under 21 different basal-bolus treatments. For each patient, a DT is fitted under one treatment and then produces glucose trajectories for the remaining 20 treatments. Our results show that prediction accuracy and decision quality disagree. A baseline that models the average patient, never seeing the individual's glucose data, has the lowest prediction accuracy of all six methods (RMSE: 63.6 mg/dL) but ranks treatments with the second-highest mean Spearman (0.766), while a black-box neural network that predicts glucose nearly twice as accurately (RMSE: 32.9 mg/dL) ranks them no better than a random ordering of the 20 treatments (0.049 Spearman). Therefore, selecting a DT based on prediction accuracy alone can prefer one that recommends treatments at chance, and decision quality must be evaluated directly rather than assumed to follow from prediction accuracy.
evaluation protocol - arxiv:2610.30576 · cs.LGBridging the Closed-World Gap: Interpretable Anomaly Detection via a Hybrid TSK-FCM Ensemble on the BETH DatasetLohith Srikanth Pentapalli, Josette Riep, Justin Ouwerkerk, Kelly Cohen
A fundamental paradox plagues the deployment of Machine Learning in cybersecurity: supervised models generalize poorly to novel, Out-of-Distribution (OoD) attacks, whereas unsupervised models often suffer from poor separability, leading to unmanageable false-alarm rates. This challenge is exemplified by the BETH dataset, a kernel-level benchmark in which the training data consists solely of benign and suspicious insider behaviors, whereas the test set consists almost entirely of novel botnet activation attacks. To bridge this detection gap, we propose a Hybrid Fuzzy Ensemble that synergizes the precision of supervised learning with the adaptability of unsupervised clustering. Our architecture integrates a gradient-optimized Takagi-Sugeno-Kang (TSK) fuzzy classifier to act as an expert on known insider threats, with a parallel Fuzzy C-Means (FCM) module to detect geometric anomalies. A critical finding of our research is that traditional fuzzy membership scores degrade in high-dimensional feature spaces due to the curse of dimensionality. We overcome this by introducing a robust Min Distance metric that exploits the raw geometric isolation of botnet activation attacks. This approach yields a state-of-the-art AUROC of 0.9951 on the BETH test set, outperforming the original Isolation Forest baseline by 17.1%. By combining near-perfect OoD detection with the intrinsic interpretability of fuzzy logic, our framework offers a viable path toward robust, transparent, and autonomous network defense.
benchmark - arxiv:2610.30573 · cs.AIDiagnosing Personalization Failures in Mobile GUI AgentsZhixin Lin, Dongliang Xu, Jungang Li, Shidong Pan +5
Unlike conventional GUI agents, personalized GUI agents require user preferences to guide sequential execution over changing GUI states. This raises two challenges: determining HOW each preference should alter execution strategy and WHEN it should take effect along the interaction trajectory. Existing benchmarks primarily evaluate overall task success, offering limited insight into these questions. We introduce PrefDiag-Bench with two components: (1) a categorization of preferences into five types, characterizing HOW they affect execution; and (2) a stage-wise diagnostic framework with 1,664 probes, tracing WHEN preferences should take effect and where failures occur. We evaluate eleven GUI agent systems, including end-to-end agents and agentic workflows, under two forms of user context: explicit user profiles and behavioral memory logs. Although existing agents achieve relatively high completion rates on non-personalized tasks, performance is substantially lower on preference-personalized tasks. Even the best-performing model, Seed-2.0-Pro, achieves only 47.17% on Process preferences under explicit user profiles. Agents struggle particularly with Ranking and Process preferences. Further analysis reveals two key capability bottlenecks: selecting relevant evidence from user history and recognizing when preferences should take effect during execution. Controlled interventions that explicitly provide relevant evidence improve end-to-end task success by up to 7.06%. PrefDiag-Bench offers a more systematic evaluation of GUI agents' personalization capabilities beyond overall task success. Code and data are available at https://github.com/Zhixin-L/PrefDiag-Bench.
memoryagentagenticagent systembenchmark - arxiv:2610.30562 · cs.LGLifetime-valid decisions by continually evolving AI analystsXiaojing Sun, Yuhan Zeng, Xiao Wang
AI agents increasingly act as analysts whose decision can often be cast as hypothesis tests. Classical tests control the Type I error rate, the probability of rejecting a true null hypothesis, for fixed procedures. However, AI analysts introduce internal randomness and can evolve continuously over time. As AI analysts evolve, the risk of approving versions with Type I error rates above the target can accumulate, and adaptive use of evaluation feedback can lead to overfitting. We introduce continual validation, which evaluates each proposed AI analyst version before deployment using simulated data under the null hypothesis, with scenarios that can be randomized and expanded over time. We prove that, regardless of how updates are chosen or how many are proposed, the lifetime probability of ever deploying a version whose Type I error rate exceeds the allowed level remains bounded. This guarantees that every subsequent decision controls the Type I error rate at the target level across these scenarios. At the same time, updates with sufficiently low Type I error rates pass validation with high probability, allowing the analyst to continuously improve its performance without sacrificing rigor. In two examples involving evolving AI analysts for bioequivalence studies and product A/B tests, Type I error rates reached 0.098 and 0.183 without continual validation, compared with at most 0.045 and 0.044 with it, while retaining 82% and 96% of the power from optimally selected valid updates.
ai agent - arxiv:2610.30557 · cs.AIGoverned Self-Evolution for Autonomous Investing: A Multi-Agent LLM System with an Invariant CoreYebo Feng, Jingyang Dai, Liangrui Pan, Nelson Lau +2
LLM agents that act on live Web data must combine heterogeneous evidence, remember what they have done and improve from outcomes without weakening their own safeguards. We study this problem in autonomous investing, where every action is irreversible and based on untrusted Web content. We present Steward, a multi-agent LLM system that draws the boundary of self-evolution in its type system. First, nine role-specialised agents decide jointly: four analysts read news feeds, SEC filings, macroeconomic series and market data and emit typed reports, from which a Portfolio Manager sets target positions that a protected Risk Manager audits. Second, a seven-layer shared memory, assembled by code from the system's own record, gives each decision its position context, the market narrative, calibration and the agent's past decisions. Third, the system improves itself by review: code grades every trade and forecast nightly and writes the grades into memory, and each quarter a reflector proposes prompt learnings that a deterministic gate applies reversibly and replay of recorded inputs verifies. All three operate around an invariant core of order arithmetic, risk rules and discipline-agent prompts that no self-edit can reach. In a five-month live deployment (102 trading days, 1,632 LLM calls), the system contained a 24.6% sector drawdown to a 0.3% loss; in replay of its recorded inputs, a single model given the same raw data traded more and lost 1.35 points against the market over five days where the joint design showed no significant loss, and no injected instruction was followed; and a 2015-2026 backtest shows the core alone lowering the maximum drawdown under all 50 random entry policies. The reflector traced the dominant failure to its root cause, and the gate applied a learning that the agent adopted on trigger evenings and on no control evening. Code: https://github.com/yebof/quant-agent
agentllm agentmulti-agent - arxiv:2610.30540 · cs.ROMotion Retargeting with Spatiotemporal Preferences for Humanoid Loco-ManipulationXiongbo Shi, Shihang Du, Zongyou Yu, Ao Zhou +5
Human motion retargeting bridges the embodiment gap between humans and humanoid robots by converting human demonstrations into reference trajectories for whole-body control and loco-manipulation. For object-interaction tasks, successful execution depends on accurately preserving body--object relations at critical stages. However, uniform per-frame iteration budgets do not account for these localized accuracy requirements, limiting the computational efficiency of retargeting. To address this limitation, we present Spatiotemporal Retargeting (STR), which uses keyframes and key body parts to allocate optimization effort. Specifically, a vision-language model selects relevant motion signals and body parts, and keyframes are identified from changes in the selected signals. Based on these selections, STR assigns additional solver iterations to keyframes and higher interaction weights to the selected parts. To further improve contact fidelity, we introduce Mechanics-Aware Contact Anchoring (MCA) to refine object-surface contact positions using an approximate object-wrench model while retaining the selected surface directions. Building on the resulting references, we develop Spatiotemporal Reinforcement Learning (STRL), which combines adaptive key-stage sampling with a contact reward based on target proximity and object contact forces for downstream control. Simulation experiments on a Unitree G1 and real-world experiments on a Unitree R1 demonstrate that STR improves geometric contact coverage and continuity at lower computational cost, while the resulting references improve downstream task completion across diverse object-interaction motions.
manipulationhumanoidwhole-body control - arxiv:2610.30537 · cs.AIPLE: Physical-Engineering's Last ExamRujun Han, Yufan Zhuang, Arijit Ray, Han Yu +9
Frontier language models now score above 90% on popular knowledge benchmarks and are increasingly deployed as agents that carry out multi-hour professional work. Yet no existing benchmark tests whether they can perform the real-world, economically valuable work of a professional physical engineer, such as modelling physics and legal code requirements across a multi-format project corpus, optimising material and installation costs, and issuing 2D/3D engineering design packages that a licensed reviewer would sign off on. Existing engineering benchmarks focus on textbook question answering, isolated tool calls or self-contained toy design problems. As a result, most of these benchmarks are already saturated by leading agents and solving them is far away from delivering real engineering design. We introduce PLE, a benchmark of 85 long-horizon, agentic tasks in professional physical engineering consisting of electrical, civil, structural and mechanical engineering, where solving a task means completing a realistic, billable engineering deliverable. PLE combines two complementary curation paths. Expert-Authored tasks are created end to end by practising engineers from a professional data provider's expert pool. Co-Authored tasks start from problem statements written by practising engineers; an adversarial Creator-Solver agent pipeline constructs each task's realistic project environment, executable grader and reference deliverables before sending them to expert engineers for validation and difficulty intensification. We evaluate a variety of frontier LLMs under seven publicly available agent harnesses. Even the strongest configuration reaches a mean score of only 50.6%, while spending 27M tokens and 63 minutes per task on average, demonstrating the substantial gap between current agents and autonomous physical engineering design.
agentagenticbenchmark - arxiv:2610.30533 · cs.CLIFC-MCQ: Automatically Generating Multiple-Choice Benchmarks for Evaluating LLM Agents on IFC Data RetrievalRabindra Lamsal, Sisi Zlatanova, Xuesong Shen
Industry Foundation Classes (IFC), the standard for building information exchange in architecture, engineering, and construction (AEC), is increasingly queried through large language models (LLMs) using natural language. Evaluating these systems at scale is difficult because free-text answers require costly human or LLM-based review. We propose IFC-MCQ, a method that automatically generates multiple-choice benchmarks from IFC files for exact option-set grading. The taxonomy covers 15 categories and 36 subtypes across four groups, with answer keys derived from the file and distractors constructed from file-derived values or predefined rules. We applied IFC-MCQ to IFC models of varying sizes and modelling characteristics to assess generation coverage and evaluated two agentic configurations on selected models using tool- and code-based retrieval. The evaluation exposed category-specific failure modes and showed that performance can vary with the source IFC model and retrieval interface. IFC-MCQ enables scalable benchmarking of LLM-based IFC retrieval in AEC workflows.
llm agentagenticbenchmark - arxiv:2610.30528 · cs.RODecision Making and Ride Comfort Evaluation Based on Lane-Change Prediction for Autonomous VehiclesMohamed Manzour, Aditya Kumar, Rubén Izquierdo, Miguel Ángel Sotelo
Predicting surrounding vehicles' lane-change intentions enables autonomous vehicles to anticipate potential hazards and respond proactively rather than relying solely on reactive maneuvers. This study investigates the benefits of integrating lane-change prediction, based on knowledge graph embeddings and Bayesian inference, into autonomous vehicle decision-making. Three driving strategies are evaluated in risky merging scenarios simulated in CARLA: reactive driving, prediction-based longitudinal action, and prediction-based lateral action. The proposed prediction model anticipates the surrounding vehicle's lane change approximately 3-4 seconds before its occurrence, allowing the ego vehicle to adjust its behavior accordingly. In the representative simulation records, reactive driving results in a collision accompanied by aggressive deceleration. In contrast, prediction-based longitudinal action initiates early braking to establish a safe gap, while prediction-based lateral action performs a lane change when a safe adjacent lane is available. Both predictive strategies avoid collisions in the evaluated records. Compared with reactive driving, peak deceleration before impact is reduced by 31.5% with longitudinal action and 91.0% with lateral action. Additionally, an ISO 2631-1-based horizontal vibration estimate is used to provide a ride comfort assessment. Overall, the results highlight the potential of lane-change prediction to support safer and smoother driving through proactive decision-making, with additional insights into ride comfort provided by vibration-based estimates.
knowledge graph - arxiv:2610.30524 · cs.LGLLM2Cert: LLM Synthesis of Certificates with SMT/Lean VerificationZhipeng Cao, Peiyang Song, Nikolaus Vertovec, Weiru Cui +5
Lyapunov functions and barrier certificates give state-space guarantees of stability and safety, but synthesizing and verifying them---especially for nonlinear, high-dimensional systems---remains an open challenge. Existing pipelines verify candidates with numerical solvers (sum-of-squares programs), interval methods, or satisfiability-modulo-theories (SMT) solvers, which often require relaxations, are conservative, or scale poorly with model dimension and nonlinearity. These limitations have also shaped benchmark design: standard test suites are dominated by low-dimensional polynomial systems chosen to suit what solvers can handle rather than what engineering applications require. We address both sides of this challenge. First, we introduce DynaCertBench, a benchmark of 105 instances: 80 non-redundant systems curated from the literature, extended with high-dimensional and control-constrained variants and new nonlinearly coupled systems, comprehensively spanning stability, region-of-attraction, forward-invariance, and safety tasks. Second, we present LLM2Cert, an LLM-driven framework that synthesizes candidate certificates and verifies them against their proof obligations: each obligation is routed to an SMT solver or directly to the Lean Proof-Repair agent. On DynaCertBench, LLM2Cert verifies 81 of 105 instances ($77.1\%$), a $17.1$ percentage point improvement over one-shot proposal without feedback. On the 81 fixed certificates, LLM2Cert verifies 22 that All-SMT leaves unresolved. Code and data are available at https://github.com/FormalCert-AI/llm2cert.
benchmark - arxiv:2610.30519 · cs.CVUnified Trajectory and 3D Pose Forecasting for Multi-Agent Sports AnalysisMarc Peral, Guillem Capellera, Luis Ferraz, Antonio Rubio +3
Forecasting future human motion requires jointly reasoning about an agent's global trajectory and 3D body pose, which are tightly coupled, particularly in team sports where multiple interacting players introduce complex social dynamics. We exploit a socio-temporal encoder that fuses trajectory and pose at the input level and models them jointly through shared factorized attention. The model is further enhanced by a per-joint loss reweighting scheme that emphasizes dynamic joints and prevents degenerate solutions, and a frequency-domain trajectory smoothing term that improves trajectory prediction while indirectly benefiting pose forecasting. To ensure fair comparison with prior joint forecasting methods, we identify and correct evaluation biases arising from overlapping sequence construction, adopting instead a non-overlapping protocol and a deterministic, oracle-free strategy for multi-future prediction. Our approach improves over prior joint forecasting methods on both WorldPose and 3DPW, with the largest gains on pose accuracy, demonstrating that the proposed joint formulation generalizes beyond the densely interactive scenarios that motivate it.
multi-agent - arxiv:2610.30518 · cs.AIProtegrity-PII Technical ReportHardeep Bassi, Hui Ma, Luís Seabra, Koowum Joshi +2
We present Protegrity-PII, a new PII decision model for structured and unstructured text. Protegrity-PII achieves a typed recall of 0.915 with a precision of 0.907 on a new benchmark we propose, PII-SCOPE. Our model processes text in 0.49 seconds per million characters at \$0.0005 per million characters on a single RTX 6000 using batched inference. Due in part to the expansive nature of the entity type taxonomy used in training, Protegrity-PII achieves superior out-of-distribution generalization when benchmarked against a selection of state of the art models and datasets.
benchmark - arxiv:2610.30504 · cs.CLRethinking the Evaluation of LLM Fact-CheckersAyana Niwa, Miyu Yamada, Kouta Nakayama, Masahiro Kaneko +2
Large language models (LLMs) are widely used to answer factual questions, but their answers can contain errors and must be fact-checked claim by claim. Yet the resulting score depends on three design decisions: which questions the claims answer, under what conditions each claim is judged, and which sources count as evidence. No existing benchmark specifies all three. We introduce JSFactCheckBench, a Japanese benchmark that specifies all three, with 1,913 claims from LLM answers to questions about misleading information that circulated on Japanese social media, each annotated for factuality at a reference time and for importance to the question, using evidence from preselected Web sources. Across six LLMs, all verify claims directly answering the question less accurately than supplementary claims, a gap that aggregate scores obscure. Scores also drop when models must locate the relevant content in longer or multiple sources and, for all four open models, when the surrounding question and response are added. These weaknesses are visible only because the benchmark records the conditions behind each label. A benchmark that leaves them implicit cannot reveal them.
benchmark - arxiv:2610.30500 · cs.ROPhysibleMDM: Enforcing Inference-Time Physical Feasibility Constraints on Motion Diffusion Models via Noise OptimizationMaximilian Stölzle, David Müller, Agon Serifi, Ruben Grandia +2
Motion Diffusion Models (MDMs) can generate diverse and visually appealing human motion; yet, the resulting trajectories frequently violate physical constraints, limiting their downstream use in simulation and control. We present PhysibleMDM, a framework that promotes physical feasibility of MDM output by optimizing at inference time the diffusion noise without retraining the generative model. We introduce a set of surrogate feasibility constraints that can be evaluated without full trajectory simulation. These constraints combine kinematic checks (e.g., ground penetration, slip, and self-collision) with dynamic feasibility tests computed via an inverse-dynamics inner loop formulated as a cone program. We solve the resulting black-box noise optimization problem using the Separable Covariance Matrix Adaptation Evolution Strategy (Sep-CMA-ES), thereby avoiding reliance on computationally expensive and discontinuous gradients. An autoregressive scheme then chains optimized segments to synthesize long motion sequences. Across 100 motion styles on an SMPL humanoid, PhysibleMDM reduces mean ground penetration, self-collision penetration, and ground slip by 75%, 92%, and 74%, respectively, relative to unoptimized diffusion output. Across 11 sequences for a custom bipedal robot, it achieves median reductions of 89.5% in per-sequence mean acceleration-reconstruction error and 79.5% in the mean inverse-dynamics residual.
humanoid - arxiv:2610.30499 · cs.AICan LLMs Improve Their Own Harness? A Controlled Study on ARC-AGI-3Jens Lehmann, Sahar Vahdati
A generally intelligent system acquires new skills efficiently without requiring humans to redesign its software scaffold for every new task. Yet language models rarely act alone. A harness is a symbolic program that orchestrates calls to the model, handling perception, memory, tool use, recovery, and inference allocation on the model's behalf. ARC-AGI-3 is an interactive benchmark in which agents must explore unfamiliar environments, figure out their goals on the fly, and keep learning as they go. On it, human-steered systems already report ceiling scores across the public games, so we know effective designs exist. We ask whether frontier models can build and improve such a harness on their own. Through a fixed coding bootstrap, each model constructs a complete task harness, acts through that harness on five development games, and revises the entire codebase twice. Every version is evaluated on twenty games that never influence adaptation. We measure task performance with Relative Human Action Efficiency (RHAE), which combines level completion with action efficiency relative to human baselines, and record dollar-denominated inference cost as a practical resource measure. Across 27 lineages, harness performance ranges from zero to near-ceiling scores. Two revisions raise mean held-out RHAE by only 2.80 points (95% interval [-1.44,7.35]). Development gains overstate held-out improvement, and reductions in inference cost can come at the expense of task performance. Human-steered systems demonstrate further headroom. These findings reveal a gap between autonomous harness construction and reliable adaptation under this two-revision protocol: even strong task performance does not ensure that subsequent revisions generalize or preserve existing capabilities.
tool usebenchmark - arxiv:2610.30495 · cs.AIFrom Prose to Properties: Testing Research Claims with ARATest for Agent-Native ScienceYifeng He, Jiachen Liu, Hao Chen
AI agents now run experiments and report findings, so research produces claims faster than researchers can check them. Each empirical claim implies a property that must hold over its cases, but it states that property only in prose. Without an explicit property, no test can decide which evidence would falsify a claim, so validation cannot keep pace with the claims agents produce. To make claims testable, we approximate a claim's semantics by abstracting its property skeleton, a parameterized relation that recurs across claims. We present ARATest, a property-based testing library that checks these skeletons as executable tests. A typed declaration binds a claim to a skeleton and fixes the claim's compared quantities, cases, scope, and numerical policy. A shared case type makes every generator supply cases in the claim's statistical unit. A three-valued semantics assigns each property one of three verdicts: holds, violated, or inconclusive. Every violation carries a replayable witness. A default generator reads the evidence that agent-native research artifacts (ARA) link to each claim, and auto-research generators dispatch an agent to run released code for fresh cases. Our DissClaimer plugin supplies that agent and runs these generators as continuous testing. Of 243 claims in 33 ARA artifacts, 54 received faithful declarations bound to evidence: 32 held, eight were violated, and 14 remained inconclusive. Two violations exposed errors in published papers that their authors confirmed. ARATest detected all 553 seeded faults without a false alarm on 19 valid controls, outperforming two LLM judges. Fresh runs exposed behaviors that reported evidence could not show. These results suggest that one fixed test can check explicit claim properties against both reported and fresh evidence.
agentai agent - arxiv:2610.30491 · cs.AIWhat the Trace Shows and What the Label Says: Auditing Behavioral Detectors in Agent EvaluationIdunnuoluwa A. Adeniji, Shilpika, Michael E. Papka
LLM agents are often evaluated by reducing execution traces to labels such as final-answer correctness, tool verification, and fault propagation. These labels are measurements produced by parsers, behavioral detectors, and harness rules, and they may not fully capture what the trace shows. We study this in a tool-use setting built on GSM8K, where agents call a calculator whose outputs are occasionally corrupted, giving clear ground truth for what the tool returned, what the model later used, and what the detectors reported. We audit three trace-level detectors across six production models and 30 problems per model. Final correctness, visible re-verification, and fault propagation can disagree in systematic ways. A key failure mode is silent tool-output override: the calculator returns a corrupted value, but the model later uses a different, often correct, value without calling the tool again, so the bad value stops propagating while a verification detector records no re-check. Human review of 90 traces agrees perfectly with the automated labels for final correctness (67/67), fault impact (19/19), and fault relevance (18/18), but only 67/81 for verification; all 14 disagreements fall into four recurring patterns. Execution limits also change the reported result: in an ablation across two models, one ranges from 21/30 to 27/30 correct as step and token budgets change, while the other stays between 27/30 and 29/30. These findings motivate measurement transparency: an automated label should come with the trace evidence behind it, whether the question the detector asks applied to that trace, whether the run was cut off by an execution limit, and any human corrections. Our trace viewer makes this information inspectable alongside the reported results.
agentllm agenttool-use - arxiv:2610.30488 · cs.AIFrom Single-Actor Environments to Multi-Actor Event Worlds: Can LLMs Be Event World Models?Juntong Ni, Zewen Liu, Bohan Wang, Lige Zhang +1
World models help agents assess the consequences of their actions before acting. Many real systems, however, involve multiple actors who make their own decisions, can evolve even when the agent takes no action, and are only partially observed through event logs. We formalize these systems as event worlds and ask whether large language models can predict future observations from the log and the agent's proposed actions. We build complementary benchmarks from human Diplomacy games and real Polymarket trades. Our hierarchical framework compares prediction with and without other actors' actions, then tests multi-step rollouts, predictions for actions never taken, and usefulness to agents. Fine-tuning and more data improve prediction when all actors' recorded actions are given. With only the agent's action, models struggle when other actors strongly affect the outcome, and larger models and longer histories help little. This gap also limits multi-step prediction and agent decisions, with models overestimating plan success. Yet imagined experience and initialization from world models can still improve policy learning. Anticipating other actors' actions remains a central challenge for event world modeling.
world modelagentbenchmark - arxiv:2610.30485 · cs.AIWhy Bother Me Again? Benchmarking Personal Agent Memory for Interaction PreferencesJialin Li, Zhenhao Chen, Hanjun Luo, Hanan Salam
Personal intelligence systems are taking on increasingly complex everyday tasks, yet repeated questions, unnecessary confirmations, and unwanted outreach can turn capable assistance into interaction burden. Reducing this burden requires agents to remember how users want to be assisted. However, existing memory benchmarks focus on factual recall and personalized answers and rarely test how agents learn and act on interaction preferences. We introduce MemPABench, an interactive environment for on-policy evaluation of memory in personal agents. MemPABench poses long-horizon tasks requiring agents to decide when to clarify, confirm, or act autonomously. Four grounded simulated users hold 14 preferences across work and personal contexts. Their feedback spans 99 sessions over four months, with five event-driven changes per user testing memory updates. We evaluate preference selection, delivered behavior, and sensitivity to context and preference change. Evaluating four memory systems across four LLM backbones, we find that current memory brings limited gains in preference selection. Retrieval-based systems (i.e. Mem0, MemOS, simple RAG) do not outperform No Memory, while File-based memory improves accuracy but applies preferences beyond relevant contexts. Our analysis shows that current memory struggles to retrieve interaction preferences and adapt to preference changes. Even with the correct preference available, overriding the agent's default interaction habits remains challenging. The environment and dataset are available at https://mempabench.com.
memoryagent memoryagentbenchmarkpolicy evaluation - arxiv:2610.30481 · cs.CVAVAtt: A Free-Viewing Eye-Tracking Dataset and Benchmark for Visual Attention on PaintingsMohamed Amine Kerkouri, Marouane Tliba, Aladine Chetouani, Alessandro Bruno
Saliency and scanpath models are developed almost exclusively on natural photographs, yet paintings are composed by artists to guide the viewer's eye and differ from photographs in texture, colour and abstraction. We introduce \avatt, a free-viewing eye-tracking dataset on \nImages{} paintings spanning sixteen art movements and five centuries, with \nScanpaths{} scanpaths and \nFixations{} fixations from \nParticipants{} participants (\spPerImageMin--\spPerImageMax{} observers per painting). For every painting we release individual scanpaths with fixation timing and eye--screen distance, fixation maps, saliency maps smoothed at one degree of visual angle, temporal saliency videos and rich art metadata. We benchmark classic and deep saliency models against human agreement, revealing substantial headroom for art-aware models, and rank 44 scanpath-similarity measures, including a new language-free family of image-encoder metrics built on ten CNN, ViT and self-supervised encoders. With controls that remove viewing biases, we show that faces are fixated \faceEnrich$\times$ more than expected and within half a second, that local structure draws the eye, that composition rules fitted from image content predict where people look, and that art style shapes how much viewers agree. Dataset, code and a Python package are available at: https://huggingface.co/datasets/Amine-CV/AVAtt.
benchmark - arxiv:2610.30480 · cs.LGFIGFLEX: A Mixed-Precision FP-INT GEMM Accelerator Co-Designed with Low-Bit LLM QuantizationFaraz Tahmasebi, Haocheng Xu, Andrew Trauger, Yomal Perera +3
Weight-only quantization keeps LLM activations in floating point and stores weights as low-bit integers, turning the dominant workload into floating-point-integer (FP-INT) general matrix multiplication. Recent quantizers reach a 2-3-bit average with mixed precision, but hardware has not kept pace: GPUs dequantize weights to FP16 before the tensor-core multiply-accumulate, and bit-parallel FP-INT accelerators run at a single weight precision, so fewer bits save memory but not computation. In this paper, we present FIGFLEX, an algorithm-hardware co-designed mixed-precision accelerator whose bit-parallel datapath scales throughput as precision falls. FIGFLEX contributes (1) a shift-add, radix-4 reformulation of FP-INT multiplication that fuses the multiply-accumulate into a unified, precision-flexible partial-product reduction and roughly halves the partial products, together with a novel INT3 optimization that reclaims the idle datapath radix-4 leaves at that precision; and (2) a mixed-precision quantizer that assigns each channel group its bit-width according to the error GPTQ cannot compensate, with a per-input-channel column scaling whose power-of-two normalization keeps the scaled activations out of the FP16 subnormal range, so that the quantized model maps directly onto FIGFLEX's datapath with the same accuracy as on a GPU. On LLaMA-2-7B, FIGFLEX attains 5.85 and 8.77 WikiText-2 perplexity at 3- and 2-bit average precision, lower than GPTQ, AWQ, SliM-LLM, AMQ, and QuIP# (without fine-tuning), while delivering 1.56x and 1.42x higher whole-chip throughput per area on the linear GEMMs of a 3-bit mixed-precision model than AxCore and FIGLUT, respectively.
memory - arxiv:2610.30478 · cs.ROExternalise the Memory: A Pre-Registered Study of Where an Embodied Agent's Experience Should LiveCheng-Yin Tu
Should an embodied agent's experience be baked into its policy weights, or kept external and retrieved at inference? A memory-augmented teacher (a VLM that retrieves lessons, with their episode images, from its own past attempts) produces trajectories on a vision-conditioned reach task; from the same trajectories we distil LoRA student policies and evaluate them under matched protocols, including one that re-attaches the teacher's frozen retrieval buffer to a fixed set of student weights. We pre-registered the hypotheses, arms, tests, and analysis rules before training and before any p-value. Distilling the memory-augmented behaviour into weights beat an action-only control by +1 pp (n.s.); the registered test proves underpowered for its own minimum effect, so we withdraw - rather than refute - the memory-in-weights claim. Attaching retrieval to a fixed set of student weights added +34 pp (z=5.15), reaching the pooled memory-augmented teacher reference (49.3%); the pre-registered protocol contrast agrees (+23 pp, z=3.36); under a pre-registered auto-reset sensitivity count the two are +36 and +29 pp. An exploratory behavioural probe with pre-registered per-arm predictions is consistent with behavioural distillation transferring the marginal correction but not the conditional, situation-specific correction that retrieval supplies. The recipe that follows is to externalise the memory: in follow-up controls registered before they ran, the same frozen buffer lifts the base model with no adapter to 48% (sensitivity 48%) and the other three adapters from 19-26% to 53-55% (sensitivity 17-21% to 50-55%), and the adapter adds nothing detectable once the buffer is attached. On a second, harder task, the retrieval advantage points the same direction but falls short of significance (+6 pp, 95% CI [-4.5, +16.4]); exploratory diagnostics tie the attenuation to memory coverage.
embodiedmemoryembodied agent - arxiv:2610.30474 · cs.LGRethinking SAE Evaluation: An Atomic Interpretable Unit Framework Reveals Hidden Polysemanticity and RedundancyRiteng Zhang, Xiaoqian Wang
Sparse autoencoders (SAEs) were originally motivated by discovering monosemantic features, a property that characterizes individual features but does not specify what the learned feature set should look like as a whole. As SAEs have increasingly been adopted for attribution, circuit discovery, and intervention, researchers have begun evaluating metrics that describe global properties of the feature set, such as sparsity, dead feature rates, reconstruction fidelity, and disentanglement, a diverse set that extends far beyond monosemanticity. However, this implicit shift beyond monosemanticity has not been formalized as a clear, well-defined objective that comes before these evaluation metrics. Without such an objective, existing metrics remain ad hoc, with their necessity unjustified. In this work, we formalize this ultimate objective by introducing a new paradigm for SAE evaluation: requiring that SAE features, and more generally dictionary-based decompositions of model representations, form an atomic interpretable unit basis, a coordinate system for semantically meaningful explanation. Under this paradigm, we organize existing and new measurements such as DAME (Dominance-Aware Monosemanticity Evaluation) as proxy metrics and introduce a comprehensive benchmarking framework. Evaluating hundreds of SAEs across multiple architectures and model families, we uncover failure modes in which some SAEs appear monosemantic under existing evaluation metrics while remaining functionally polysemantic: the most dominant features, those contributing most to model computation, are systematically less interpretable and more redundant. These failure modes warrant attention from SAE researchers as targets for improved training and architectures, and motivate dominance-aware evaluation as a standard check for practitioners before SAE features are deployed for attribution, circuit tracing, or intervention.
benchmark - arxiv:2610.30469 · cs.LGLearning to Incentivise: Using Reinforcement Learning for Sustainable Urban MobilityGermán Pardo-González, Shaghayegh Vosough, Katerina Papadaki, Claudio Roncoli
Traffic management has traditionally focused on toll-based and road pricing solutions. However, road pricing often raises concerns about accessibility and public dissatisfaction, leading to its prohibition in some places, such as Finland. This study optimises the dynamic allocation of incentives to drivers, encouraging them to reroute onto alternative (potentially longer) paths to achieve greater societal benefit, reduced total travel time (TTT) and total emissions in the transportation network, contributing to sustainable urban mobility. We employ a multi-agent reinforcement learning approach to dynamically assign incentives to drivers to reduce both TTT and emissions, with travel times estimated using traffic simulation software. We demonstrate that, with an unlimited budget and an objective of minimising travel time, the incentive scheme reduces TTT by 16%, compared to the dynamic User-Equilibrium (UE) with a budget equivalent to about 11% of the UE total time. When the goal is to minimise emissions, a 9% reduction in CO2 emissions is observed under an unlimited budget. We demonstrate a critical trade-off: minimising TTT leads to an increase in emissions, while prioritising emission reductions raises TTT. However, with the right combination of weights in the multi-objective function, both TTT and total emissions are improved beyond the baseline.
multi-agent - arxiv:2610.30465 · cs.LGFlow-based Spectral Kernel Learning for Nonstationary AttentionZi Yang, Ying Li, Alejandro Lancho, Darío C. Larese +2
Kernelized attention provides a useful perspective for understanding self-attention as a similarity-based aggregation mechanism, but existing approaches often rely on fixed or stationary kernels, limiting their ability to adapt the similarity structure to complex token interactions. In this paper, we propose flow-based spectral kernel attention (FSKA), a framework for learning expressive nonstationary attention kernels. FSKA formulates kernelized attention through spectral density learning and parameterizes the resulting bivariate spectral density with normalizing flows. This design enables flexible and regularizable kernel learning while remaining compatible with scalable linear-attention computation. We further show that expressive kernel learning enables a simplified attention design with removed query--key projections and fixed orthogonal value projections, resulting in a more parameter-efficient architecture. Controlled experiments on classification, ablation, and scalability profiling show that FSKA achieves competitive or improved performance over existing attention mechanisms, while reducing projection-related trainable parameters by approximately $40\%$--$47\%$ compared with most baselines and maintaining favorable linear-attention memory scaling. Code Available: https://github.com/ziyang18/FSKA.
memory - arxiv:2610.30450 · cs.LGSame Beliefs, Different Policies: Learning Reusable Representations for Early Stopping in LLM ReasoningJingwei Ni, Dmitry Abulkhanov, Vladislav Smirnov, Markus Leippold +4
Reasoning LLMs often continue reasoning after reaching a correct answer, wasting computation and sometimes replacing it with an incorrect one. Deciding when to stop is a metareasoning problem: the model must weigh what further reasoning could achieve against its cost, given what it can already answer. Direct utility training learns a stopping representation at a chosen token price; supporting new prices can require additional controller training. We introduce BELT (BELief-based Termination), which separates outcome prediction from stopping control. It learns beliefs about current and future answers from a frozen LLM's hidden states, then fits a linear stopping policy on the frozen features. The prediction targets are independent of token price, so one probe supports different accuracy-cost trade-offs through CPU policy refitting or threshold adjustment. Averaged over five out-of-distribution (OOD) math benchmarks, BELT reduces generated tokens by 41% on Qwen3-1.7B and 51% on OLMo-3-7B-Think without loss of accuracy. The same math-trained BELT transfers to coding and agentic tasks without training on code or agentic data. In all three domains, its accuracy-cost trade-offs are competitive with or better than those of end-to-end controllers. On OLMo3, its probe has 21.2x fewer parameters than one checkpoint of the strongest baseline, OS-Pruner, and needs 55x fewer training GPU-hours than OS-Pruner's five price-specific checkpoints.
agenticbenchmark - arxiv:2610.30446 · eess.SYStrategic Arbitrage in Electricity Markets: A Multi-Agent Policy Optimization FrameworkEli Brock, Srikar Babu Gadipudi, Somayeh Sojoudi, Javad Lavaei
Grid-scale storage and load aggregators are growing quickly in wholesale electricity markets. If highly consolidated, these new participants may distort the competitive market equilibrium through strategic withholding behavior. Traditional market power mitigation techniques are ill-suited to arbitrage-based participants. We provide a multi-agent policy optimization model to study game-theoretic equilibria for real-time markets with consolidated storage and/or aggregators. The model is designed to facilitate natural first-order equilibrium-seeking update rules. We analyze the theoretical properties of different supply functions and their implications on the convergence of the gradient dynamics. Finally, we train equilibrium policies for a modified version of the IEEE RTS-GMLC benchmark across different levels of storage fleet consolidation, to demonstrate the potential of the framework as a tool to characterize and quantify strategic arbitrage.
multi-agentbenchmark - arxiv:2610.30445 · cs.AIA Three-Pillar Governance Framework for Safe and Trustworthy AI AgentsEdward C. Cheng, Jeshua Cheng, Alice Siu
AI agents capable of autonomous decision-making introduce significant challenges related to safety, alignment, accountability, and human oversight. Existing Human-in-the-Loop (HITL) and responsible AI approaches provide important foundations, but many remain fragmented across system design, governance, and operational deployment. This paper presents a Three-Pillar (3P) model and framework for developing safe and trustworthy AI agents through transparency, accountability, and trustworthiness. The framework defines an evolutionary path in which increasing agent autonomy is introduced through progressive validation, structured human oversight, and auditable operational controls. We further describe how the model supports agent lifecycle monitoring, decision traceability, escalation management, and human-agent collaboration within enterprise agentic environments. By integrating oversight principles with operational mechanisms, the proposed framework provides a practical foundation for deploying AI agents that remain aligned with human intent while enabling controlled progression toward higher autonomy.
agentai agentagentichuman-in-the-loop - arxiv:2610.30443 · cs.AISpecMAGIC: Specification-Guided Multi-Task Generalization in Multi-Agent Reinforcement LearningZijian Guo, İlker Işık, H. M. Sabbir Ahmad, Wenchao Li
Multi-task generalization is an important yet underexplored problem in multi-agent reinforcement learning (MARL). Many real-world applications require multi-agent systems to generalize across complex, temporally extended tasks with diverse inter-agent coordination requirements. We introduce a specification language that combines temporal-logic task specifications with operators for expressing such requirements. We propose SpecMAGIC, a hierarchical framework that leverages the automaton structure of task specifications for subtask decomposition. A centralized high-level planner identifies and allocates compatible subtask sequences across agents, while decentralized low-level policies execute individual subtasks. This enables zero-shot generalization to new tasks by composing reusable subtask policies according to new specifications, without retraining. We establish soundness and probabilistic completeness of the high-level planner. Experiments on multi-agent navigation tasks show that SpecMAGIC substantially outperforms existing approaches on unseen task specifications and maintains strong performance as the number of agents increases.
multi-agentagent system - arxiv:2610.30438 · cs.ROTactile-Based Robotic Manipulation: From Mechanics-Driven to Data-Driven PerspectivesJingyi Xiang, Yixuan Wang, Wenzhen Yuan
Tactile sensing provides robots with direct access to physical interactions at the contact interface, making it a critical modality for robust and dexterous manipulation. Over the past several decades, tactile-based manipulation research has evolved from mechanics-driven approaches, which explicitly interpret tactile signals through contact mechanics and task-specific reasoning, to data-driven approaches that learn tactile representations and manipulation policies directly from data. This paper reviews the development of tactile-based manipulation from these two complementary perspectives. We first introduce the fundamentals of tactile sensing, and then review, from a mechanics-driven perspective, how those signals form manipulation-relevant state representations and subsequently influence the robot action. We then examine the recent advances in data-driven tactile manipulation, covering imitation learning, reinforcement learning, vision-language-action models, and predictive models. Finally, we discuss open challenges and future opportunities toward scalable, generalizable, and robust tactile-enabled robotic systems.
vision-language-actionmanipulationdexteroustactile - arxiv:2610.30436 · cs.AISecure Agentic AI and Multimodal Intelligence in 6G Communication SystemsSatyam Gaba, Devanshu Motwani, Siva Sai, Vineet Suri +2
Sixth-generation (6G) communication systems are expected to support immersive, intent-driven, and latency-critical services through tight integration of multimodal intelligence, edge computing, and autonomous network control. In this setting, agentic AI systems can reason over heterogeneous inputs, use tools and network functions, coordinate across distributed entities, and adapt to dynamic environments. These capabilities make agentic multimodal intelligence promising for 6G applications such as semantic communications, digital twins, and zero-touch network management, but they also introduce security and privacy risks that extend beyond conventional AI and network security models. Vulnerabilities can propagate across multimodal data pipelines, memory, tools, communication interfaces, and inter-agent coordination loops. We survey multimodal intelligence and agentic AI for 6G cross-layered communication systems from the perspectives of privacy and security. We present a 6G-oriented system view spanning perception, memory, tool use, and multi-agent orchestration; analyze threat models and attack surfaces across sensing, cognition, action, memory, and communication layers; organize existing attacks into a structured taxonomy; and review mitigation strategies including prompt and memory defenses, inter-agent trust mechanisms, secure orchestration, and privacy-preserving learning. We conclude by outlining open research directions for trustworthy multimodal multi-agent 6G systems.
memorymulti-agentagentictool use - arxiv:2610.30435 · cs.AISelf-Assessment Reinforcement Learning for Search-State Awareness in Retrieval-Augmented Large Language ModelsTaewhoo Lee, Minju Song, Jungwoo Park, Chanwoong Yoon +5
Recent advances in search agents have enabled large language models to tackle complex tasks by retrieving and synthesizing external information. However, existing search agents often lack search-state awareness, i.e., the ability to recognize what information has been obtained and what remains missing, leading to hallucinated reasoning and prematurely terminated search. To address this limitation, we propose Self-Assessment Reinforcement Learning (SARL), a reinforcement learning framework in which models learn to recognize their own search state during training and improve their search behavior accordingly, without requiring additional assessment at test time. SARL achieves this through a dual reward design that captures the corrective effect of each assessment while penalizing excessive reliance on it, jointly encouraging effective and efficient search trajectories. Under this setup, SARL achieves a 24.3% relative improvement over the strongest baseline, with average gains of 35.9% on multi-hop reasoning and 27.9% on deep research benchmarks against the strongest per-benchmark baselines. SARL further surpasses methods built on auxiliary supervision or multi-agent coordination, despite operating as a standalone RL framework with no inference-time scaffolding. Beyond final performance, SARL maintains training stability under extended search horizons, a regime where existing methods consistently degrade, enabling sustained gains as task complexity grows.
retrieval-augmentedmulti-agentbenchmark - arxiv:2610.30434 · eess.SYPlug-in Estimation Method for Topology and Admittance Estimation in Electric GridsNorak Rin, Iman Shames, Ian R. Petersen, Nanduni I. Nimalsiri +1
This paper proposes a scalable method for topology and admittance estimation in electric power grids. Our proposed plug-in estimation method incorporates precise phasor measurement unit data to estimate planned or unplanned grid topology and admittance parameter changes. Our plug-in estimation method decouples the problem into two stages: (i) constructing and averaging voltage coefficient matrices and current vectors from phasor data, and (ii) estimating admittance parameters via a least-squares regression approach with thresholding to recover the network topology. Rigidity matrix theory underpins the design of our plug-in estimation method --- to determine the data requirements (e.g., necessary and sufficient conditions) for accurate estimation of the grid topology and admittance parameters. By means of numerical simulations on the IEEE 13-node, 123-node, and European low voltage test feeders, we demonstrate the scalability of our plug-in estimation method and benchmark the computational time against a structured total least squares method. Our simulation results highlight the practicality of the proposed method for topology and admittance estimation in large-scale electric networks, including electric grids with partial topology information.
benchmark - arxiv:2610.30429 · cs.AIRoleFlow: Self-Evolving Multi-Role Agentic Collaboration via Structural AdaptationYi Ge, Pan Wang, Xuehao Cui, Zian Zeng +3
Large language model (LLM) based multi-agent systems often rely on predefined collaboration structures, limiting their adaptability to evolving task requirements. Although recent approaches enable dynamic orchestration, trajectory-level rewards they apply assign one outcome to each decision, ofen struggling to identify the contributions of individual structural changes. We propose RoleFlow, a self-evolving multi-role agentic system that refines collaboration structures through feedback-driven atomic edits and on-demand role synthesis. To make structural adaptation learnable, we introduce Single-Edit Counterfactual Policy Optimization (SECPO), which evaluates each edit against its no-edit counterpart at the same chain state and assigns targeted credit to edit-specific tokens. Across seven benchmarks spanning four domains, RoleFlow achieves the optimal average performance, surpassing the strongest competitor by 4.7 points. It leads on all three multi-hop QA benchmarks by 3.2-7.5 F1 points and also shows supremacy on mathematical reasoning, web interaction, and code generation. Further analyses demonstrate cross-benchmark transferability, training-data efficiency, and cross-executor generalization.
multi-agentagenticagent systemself-evolvingbenchmark - arxiv:2610.30428 · eess.SYComputationally Transparent Power Flexibility Aggregation via Directional Support EvaluationMingyu Zhai, Liang Du, Yue Chen
The growing penetration of distributed energy resources (DERs) creates new opportunities for improving resource adequacy, operational flexibility, and power-system reliability. However, efficiently characterizing the aggregate flexibility of heterogeneous DER portfolios remains challenging because of high dimensionality, diverse operating dynamics, and spatio-temporal coupling constraints. This paper proposes a computationally transparent framework that formulates aggregate flexibility characterization as a sequence of linear programming (LP)-based directional support evaluations. The framework accommodates heterogeneous DER technologies and configurable device-, feeder-, and portfolio-level constraints. The convex hull of the resulting boundary trajectories provides a feasible inner approximation of the coupled aggregate power flexibility region. Theoretical analysis establishes feasibility and non-overestimation guarantees and characterizes the computational dependence on portfolio size, scheduling horizon, and direction budget. In the small-scale case, the proposed method produces tighter projected regions than box, zonotope, and Chebyshev inner-ball benchmarks and matches a dense directional reference within numerical tolerance using 16--64 construction directions. Large-scale simulations with up to 2000 DERs demonstrate approximately linear empirical runtime scaling with the portfolio size and direction budget over the tested ranges. The results establish a scalable and theoretically grounded approach for aggregate DER flexibility characterization.
benchmark - arxiv:2610.30422 · cs.LGMemory-Augmented Online Test-Time Adaptation for EEG Motor Imagery via Perturbation Consistency and Confidence CalibrationKunkun Zhang, Kehao Wu, Zhitao Chen, Haigen Hu +1
Recently, electroencephalography (EEG)-based brain--computer interfaces (BCIs) have gained traction as portable neural measurement systems. However, the practical deployment of these instruments is often hindered by nonstationarity and cross-subject distribution shifts, which lead to significant intersubject inconsistency. Existing adaptation methods remain impractical for real-time measurement scenarios, as they typically require offline access to source or aggregated target data. To address these challenges, this article proposes memory-augmented online test-time adaptation (OTTA) with perturbation consistency and confidence calibration (MOPC), a source-free framework designed to enhance the reliability of BCI instrumentation. MOPC enables the measurement system to update its decoding parameters on the fly during the inference stage, facilitating single-pass adaptation without accessing sensitive source data. We introduce random-mask perturbations to enforce prediction consistency against inherent EEG signal noise and acquisition artifacts. Furthermore, a confidence-calibrated reweighting loss is designed to integrate feature-space similarity, effectively mitigating unreliable pseudolabels and class imbalance during continuous signal acquisition. To ensure system stability and prevent the catastrophic forgetting of established neural patterns, a lightweight memory module is incorporated to regularize parameter updates within a tight latency budget. Evaluated on the BCI IV IIb and OpenBMI datasets, MOPC achieves state-of-the-art performance, outperforming prior methods by 2.22~\% and 3.67~\%, respectively, while keeping the per-trial adaptation latency to approximately 40--60 ms. This study provides a robust, recalibration-free solution for real-time EEG decoding, marking a significant advancement in the precision and usability of practical BCI measurement instruments.
memorymemory module - arxiv:2610.30418 · cs.AIAgent Mosaic: Recognizing Confidentiality Leaks in Agentic LLM SystemsDaniel Ruiz, Luca Luceri, Emilio Ferrara
Modern LLM agents can retrieve information from accessible documents, and then disclose a sensitive conclusion assembled from them. This mosaic effect creates a critical risk for agents operating over confidential data. We quantify its realization with 30 synthetic cybersecurity, medical, and military examples, yielding 10,800 final answers from three LLM agents in controlled simulation and 1,440 answers from live agents in two open-source harnesses. Even under benign queries designed to be answerable from one document, GPT-5.6-Sol, the best performing model tested, discloses forbidden composites in 7.08% of answers (95% CI: 3.75-10.67%) in simulation, with worse disclosure rates in live harnesses. Adding visible classification headers and explicit aggregation rules reduces pooled disclosure rates substantially, yet unsafe disclosures persist in 32.37% of simulated answers (95% CI: 28.33-36.19%) across all intents despite these safeguards. Furthermore, increased refusals do not guarantee confidentiality, since significant refusals reveal the prohibited conclusion regardless. These results point to a troubling conclusion: sensitive data leaks via mosaic-style events are an observed confidentiality failure mode in the evaluated settings at non-negligible rates. They motivate stronger controls on what agents may infer and disclose in sensitive data environments with mixed-clearance users.
llm agentagentic - arxiv:2610.30412 · cs.CLSIM-UT: A User Simulation based Evaluation Framework for Medical Triage ChatbotsDeepali Mittal, Dheeraj Mundhra, Yulia Sidorova, Tersia de Kock +2
LLM-based chat agents are increasingly used for complex, multi-turn, multilingual medical triage in low- and middle-income countries (LMICs), yet evaluation methods focus on single-turn, English-only benchmarks that primarily test the chatbots' knowledge. In practice, users are often confused, overwhelmed, or emotionally vulnerable, giving unclear, misspelled single-word answers or long messages in regional dialects, and the context of a conversation can shift rapidly. We propose SIM-UT (Simulation of User Behaviour for Triage), a multi-turn, simulation-based evaluation method that identifies ecologically valid, contextually relevant failure modes of LLM-based triage chat agents. Using SIM-UT, we evaluated three real-world WhatsApp-based health chatbots: a respiratory screening programme in India, a hearing health support programme in South Africa, and a diabetes and lifestyle coaching agent in Kenya. We compared a medically fine-tuned open model (MedGemma) against general-purpose frontier models (ChatGPT and Claude) across 5,000+ multilingual simulated conversation turns grounded in the personas, constraints, and communication patterns of the real user populations, over 2-4 iterative rounds. A key failure mode emerged in both model families: de-escalation risk. The agents initially identified medical emergencies correctly, but when the user minimised the symptoms without introducing new clinical information, they de-escalated, exhibiting clinically dangerous sycophancy, for example agreeing that vaginal bleeding in pregnancy was "not a problem" or that decreased fetal movements at 32 weeks "can be normal". We recommend user simulation-based behavioural evaluation, grounded in real user context and covering adversarial and edge cases, as standard practice for triage chatbots in production, so that safeguards can be added before deployment or in subsequent iterations.
agentbenchmarkevaluation framework - arxiv:2610.30405 · cs.CVVisCurate: Evaluating Behavioral Equivalence of Visual SkillsMd Farhan Ishraq, Md Farhan Ishmam, Reaz Hassan Joarder, Md Hasanul Kabir
Modern vision-based agents have shifted toward skill-based problem-solving, where a skill is an executable image-to-image operation, e.g., cropping, thresholding, or edge detection. As these skill libraries grow, they accumulate duplicated, mislabeled, and defective skills. The signals normally used to curate them (a skill's name, description, or source code) say little about how the skill visually changes the image. We present VisCurate, a framework and benchmark that curates visual skill libraries from observed execution rather than from text. VisCurate runs each candidate skill pair on a shared probe battery and assigns one of six behavioral relations, each tied to a library edit that is safe to make. The benchmark injects seven defect types into a clean library at controlled rates, producing 300 reproducible instances with ground-truth relations and repairs. While text baselines accept up to five of six false merges on hard negatives, our verifier accepts none and recovers 51.2% of the pairs that are truly mergeable. On two real, uncurated libraries, VisCurate finds that 92% of the skills a package shares with its downstream copy are redundant, while execution confirms only 10 of 17 published equivalence claims. The curation problem also remains open, as the strongest of the nine automated curators recovers only about half of the repairs. We emphasize that visual skill libraries should be curated by behavior, not by description.
benchmark - arxiv:2610.30394 · cs.LGEscaping the Phylogenetic Trap: Persistent Homology Enables Cross-Species Ion Channel Classification via Multimodal Graph Neural NetworksArezoo Bybordi, Favour Achimba, Mariam Gelashvili, Jessy Ramirez +3
Protein classifiers are highly sensitive to evaluation protocol: random train/test splits induce phylogenetic leakage, allowing models to exploit taxon-specific correlations and substantially overestimate out-of-distribution performance. This issue is particularly acute for venom compound classification by ion-channel target (Ca2+, K+, Na+), where reported accuracies of 80-90% do not reflect generalization to unseen taxa. We formulate cross-species prediction as a structured distribution shift in which taxonomic identity acts as a confounder, requiring representations that capture structural features invariant across phylogeny. Using a shared multimodal backbone that fuses ESM-2 sequence embeddings with a disulfide-aware graph attention network, we systematically compare five structural inductive biases (spectral, Riemannian, SE(3)-invariant, statistical-manifold, and persistent-homology representations) under a leave-one-taxon-out evaluation protocol on a benchmark of 1,373 ColabFold-predicted venom compounds spanning five taxa. The persistent-homology representation summarizes backbone geometry through topological descriptors that are robust to local structural perturbations, providing an inductive bias toward phylogenetically invariant learning. Under conventional i.i.d. splits, all models achieve similar accuracy (81-86%), masking differences in inductive bias. Under taxonomic shift, this parity collapses: a sequence-only baseline drops to approximately 30%, the remaining geometric priors cluster at 54-55%, while our persistent-homology model (PH-GNN) achieves 63.4%, outperforming the next-best method by over 8 percentage points. These results demonstrate that random-split evaluation obscures meaningful differences in representation quality and identify topological representations as an effective inductive bias for learning under structured biological distribution shift.
benchmarkevaluation protocol - arxiv:2610.30391 · cs.LGSoft Ontology Priors over Latent Concept Posteriors for Weakly Supervised Neuro-Symbolic LearningTengwei Song, Robert Hoehndorf
Neuro-symbolic (NeSy) predictors that supervise only the task label suffer from reasoning shortcuts: distinct concept mappings achieve the same maximum likelihood, leaving the recovered concepts unidentifiable and unsafe for downstream reuse. We attack this failure by injecting a soft ontology prior over the latent concepts into the training objective and compiling it together with the task constraint into a smooth, decomposable probabilistic circuit. Under independent Bernoulli perception, the normalisers of the resulting structured concept posterior reduce to weighted model counts that are linear in the compiled circuit size, because the neural leaf weights fold into the circuit before evaluation. The soft compilation strategy places expert axioms in the circuit's weights rather than its support (task-structural axioms guaranteed by the data-generating process stay hard), so noisy expert axioms only redistribute prior mass and cannot empty the support. We characterise how the prior reweights label-restricted shortcut mass through differences in ontology violation energy, show that soft priors incur bounded log-density perturbations under bounded axiom noise, and prove that the hard-support counterpart can be made degenerate by a single corrupted axiom. We validate these predictions on an audited six-concept view of the real-world BDD-OIA driving benchmark and on controlled NeSy tasks.
benchmark - arxiv:2610.30388 · cs.LGA Transformer-based Approach for Real-time Multi-agent Path Planning with Varying Initial/Final ConditionsQian Zhang, Shanqing Liu, Jerome Darbon, George Karniadakis
We address the problem of real-time multi-agent path planning under prescribed initial and final configurations, formulated as a high-dimensional optimal control problem subject to inter-agent collision avoidance and obstacle constraints. Classical solvers based on the Hamilton--Jacobi--Bellman equation or Pontryagin's maximum principle (PMP) become computationally intractable as the number of agents grows, due to the curse of dimensionality and the non-convexity introduced by interaction constraints. We propose OCFormer, a transformer-based neural network that parameterizes the inverse of a coordinate transformation mapping a simple, analytically tractable latent space, in which agents evolve independently under a decoupled Hamiltonian, to the physical space of the coupled, interacting system. The model is trained by minimizing a physics-informed loss comprising the Hamiltonian ODE residual from the PMP and a boundary matching term, without requiring pre-computed trajectory data. The self-attention mechanism endows OCFormer with permutation equivariance over agents, a structural symmetry that prior symplectic network approaches do not natively enforce. When trained on a distribution of boundary conditions, the amortized OCFormer produces near-optimal trajectories for unseen initial and final configurations in a single forward pass ($\sim$220ms), enabling real-time inference without retraining. We further introduce a lightweight transfer learning framework that adapts a pretrained OCFormer to out-of-distribution perturbations in seconds via L-BFGS fine-tuning. Numerical experiments on two- and three-dimensional multi-agent navigation with up to 128 agents demonstrate the scalability and efficiency of the proposed approach.
multi-agent - arxiv:2610.30379 · cs.AIMemPlan: Memory-Conditioned PDDL Planning for Partially Observable Text EnvironmentsTengwei Song, Robert Hoehndorf
LLM agents are flexible in language interaction, but using them as long-horizon planners is costly and prone to hallucinated or invalid actions. This paper targets a different role for LLMs: constrained bridges between language and symbolic planning. We propose MemPlan, a memory-conditioned PDDL planning framework for partially observable interactive text environments. MemPlan keeps the PDDL domain fixed and reconstructs the PDDL problem at each step from parsed observations, candidate hypotheses, and interaction memory. Positive graph memory assigns relevance-based action costs to candidate hypotheses for missing objects, locations, or conditions, guiding the planner toward more plausible checks. Negative memory records execution-refuted assumptions and compiles them into exclusion constraints, preventing repeated attempts at candidates already contradicted by feedback. A unified schema-conditioned language--symbol interface, trained offline with supervised fine-tuning, connects text environments to the planning loop through observation-to-fact parsing and symbolic-action-to-command grounding. Experiments on TextWorld, ALFWorld, and Robotouille show that MemPlan improves task completion and step efficiency over language-only and PDDL-aware baselines, while substantially reducing online token cost compared with LLM-planning baselines such as ReAct. The same fine-tuned interface is also reused across TextWorld task types without task-specific fine-tuning. Our code and data are available at: https://github.com/tewiSong/MemPlan.
memoryllm agent - arxiv:2610.30355 · cs.CLCiteAnchor: Source-Aware Grounding for Hallucinated Citation Detection and ResolutionChenyi Tong, Ping Nie, Yu Zhang
Hallucinated citations make it harder to trace sources and verify the evidence behind scientific claims. Detecting them is difficult because bibliographic records can be incomplete or differ across source types and publication versions. A plausible match alone neither validates the citation nor establishes its underlying source. We introduce CiteResolve, a benchmark of 423 real-world hallucinated citations annotated for source recoverability. We further propose CiteAnchor, a source-aware grounding framework that adapts retrieval and field-level verification to each source type and separately determines citation validity and whether a unique source can be recovered. On an existing detection benchmark, CiteAnchor achieves 0.841 F1 with GPT-5.6 Luna and 0.784 with an open-weight backbone, outperforming the baseline with the same backbones by 7.9 and 36.8 points. On CiteResolve, it achieves 74.7% and 70.7% end-to-end task success, outperforming the corresponding web search baselines by 16.3 and 65.3 percentage points. Our code and data are available at https://github.com/oldwizard7/CiteANCHOR_code.
benchmark - arxiv:2610.30351 · cs.LGRoute-and-Generate: KL-Regularized Router Improvement for Reasoning Routing in LLMsTien-Phat Nguyen, Truong Nguyen, Thanh-Thien Le, Hoang Phan +3
Large language models often apply complex reasoning indiscriminately, leading to overthinking on simple tasks and underthinking on harder ones. We present Route-and-Generate (Rouge), a hierarchical training framework that enables LLMs to adaptively decide when and how much reasoning is required for each input. Our approach decouples reasoning effort allocation from answer generation by introducing a routing policy that allocates a fixed sampling budget across multiple reasoning formats, while the base model is optimized with GRPO using per-sample rewards. To promote efficient reasoning, we employ a rank-based reward that favors shorter correct solutions, and update the routing policy via a closed-form KL-regularized policy improvement step distilled into a neural router. This design encourages concise answers on easy problems while reserving longer reasoning for harder ones, achieving improved accuracy-efficiency trade-offs across multiple reasoning benchmarks.
benchmark - arxiv:2610.30349 · cs.LGConsistency Is Not Validity: Self-Rewarded Moral Argumentation Makes Language Models Decisive but PliableMarius De Kuthy Meurers, Aaron Parisi, Bernd Bohnet
LLMs are increasingly consulted for moral judgment. We aim to improve performance on such tasks by rewarding good arguments, on the premise that arguing well increases the likelihood of judging well. We train a reinforcement learning policy initialized from a state-of-the-art language model in a self-judged tournament with masked thinking: the generated arguments are judged pairwise on argument quality by the policy itself. As an empirical basis, we extract authentic dilemmas from classic novels, using a logic solver to verify that each constitutes a conflict. We show that our judge rates argument quality like human annotators, and the policy then adopts the language register that it rewards. Testing on the external benchmarks MoralChoice, ELEPHANT, and MORABLES, we find that the policy becomes more decisive and invariant to the order in which options are presented, but sycophancy and reframing remain a challenge to validity. Two ablation studies with synthetic dilemmas and unmasked thinking match the reordering gains, but show a weaker register shift and higher sycophancy. We conclude that consistency is not validity: consistency under reordering can be optimized without improving the moral judgment it is meant to signal.
benchmark - arxiv:2610.30347 · cs.CVFiLM-CAM: Keyed Feature Modulation for Conditional-Access WatermarkingYash Kulthe, Vishal Asnani, Shruti Agarwal, Andrew Gilbert +1
We propose FiLM-CAM; a conditional-access image watermarking approach that enforces authorized encoding, decoding, and removal through keyed feature modulation. A secret key conditions a Feature-wise Linear Modulation (FiLM) mechanism that transforms the internal representation of the watermark signal before embedding, thereby coupling the payload to the key at the feature, rather than pixel level. The same key must be presented to condition the decoder to correctly recover the payload embedded in an image. To enforce conditional access, we train the system to maximize the performance gap between authorized and unauthorized use, ensuring reliable extraction under the correct key while inducing failure under incorrect keys. In addition, we design our encoder to estimate the watermark residual invariant to the presence of pre-existing watermarks. This design enables conditional watermark removal via simple re-encoding and subtraction of the residual using iterative refinement, contingent on knowledge of the secret key. We evaluate FiLM-CAM across multiple watermarking encoder-decoder backbones, demonstrating that keyed feature modulation serves as an effective, architecture-agnostic conditional access module (CAM) for image watermarking.
iterative refinement - arxiv:2610.30344 · cs.LGAQL-Lite: Enabling Efficient Knowledge Acquisition for Arabic LLMs via Data-Centric Continual PretrainingHoussam Eddine Boukhalfa, Mourad Bouache, Mourad Dahmane
Training Arabic large language models faces challenges from noisy web data, dialectal variation, and the high cost of continual pretraining (CPT). We introduce AQL-Lite (AQL: Arabic Quality Language), a time- and resource-efficient framework for adapting a multilingual LLM to Modern Standard Arabic (MSA). AQL-Lite couples a quality-first curation pipeline (structural cleaning, heuristic and model-based filtering, MSA selection, and deduplication) with a partial-parameter CPT recipe that trains RS-LoRA adapters across the network together with the last eight transformer blocks and the embedding matrix, under layer-wise learning rates. At a matched 1B-token budget, CPT on AQL-curated data raises a four-benchmark Arabic mean by 3.51 points over CPT on raw web text, which yields no gain over the base model. In a controlled 5B-token comparison, AQL-Lite recovers 94% of the gain of full-parameter CPT (mean 73.19 vs. 73.58) in 52% of the wall-clock time and outperforms adapter-only baselines on every benchmark, and a component ablation shows that adapters, unfrozen blocks, and embedding training each contribute. Scaled to 40B MSA tokens and instruction-tuned on one million examples, the resulting 12B model scores 75.3 on the seven-benchmark Open Arabic LLM Leaderboard v2 (OALL v2) suite. These results suggest that partial-parameter CPT on a carefully curated corpus is a practical route to efficient MSA adaptation.
benchmarkleaderboard - arxiv:2610.30340 · cs.CVFoundation4D: Towards a unified perception and reconstruction model for autonomous vehiclesNVIDIA, :, Sérgio Agostinho, Maximilian Apodaca +24
We introduce Foundation4D, a unified feed-forward model for geometry estimation, 3D detection, semantic occupancy, and Gaussian reconstruction from multi-camera driving video. A shared backbone uses factorized spatiotemporal attention to efficiently combine observations across cameras and time. Our video inference pipeline uses pose graph optimization to refine camera trajectories and assembles object tracks across input windows. We also develop a data annotation pipeline and release PAI-Foundation4D, a benchmark with dense depth and semantic occupancy labels for 1,253 driving clips. With benchmark-specific fine-tuning, Foundation4D achieves the lowest depth AbsRel on Waymo and nuScenes, the highest detection mAP on nuScenes, and the highest occupancy mIoU on OpenOccupancy among the compared methods. Our task-combination study shows benefits for both detection and geometry: co-training detection with geometry and occupancy improves top-class mAP by 1.2 points over detection-only training, while co-training geometry with detection reduces camera-to-rig translation error by $13.4\%$ relative to geometry-only training. We further use the predicted geometry and trajectories as priors for lidarless NuRec, supporting reconstruction of driving scenes for simulation.
benchmark - arxiv:2610.30339 · cs.LGSpike-Aware Gradient-Based Mixed-Precision Quantization of Spiking Neural Networks for Vision ApplicationsRajshekhar Rakshit, Devgokul B., Rakshitha Kalkura, Gopalakrishnan Srinivasan
In recent years, Spiking Neural Networks (SNNs) have gained massive attention as an energy-efficient neural computing paradigm due to their event-driven and sparse computing capability. However, maintaining accuracy under low-precision constraints, necessary for practical deployment, has been largely unexplored for SNN models. In addition, existing Hessian-based approaches for quantization, such as HAWQ-V2, incur significant memory and computational overhead due to temporal unrolling across multiple SNN timesteps. We propose a spike-aware gradient-based mixed-precision quantization framework for SNNs that adaptively assigns bit-widths using first-order gradient information, avoiding costly second-order Hessian computations. We evaluate our method across various SNN-based architectures, including VGG, ResNet, and Vision Transformer models obtained from both ANN-to-SNN conversion and direct training frameworks, on CIFAR10/100, ImageNet, and DVS-CIFAR-10 datasets. Notably, on ResNet-34 evaluated on ImageNet, our method achieves 73.30% accuracy at 4.78$\times$ compression, which is comparable to that provided by HAWQ-V2, while reducing memory consumption by 2.31$\times$ and bit-allocation runtime by 100$\times$. We also present an energy analysis for mixed-precision SNNs showing that the proposed quantization strategy enables energy-efficient deployment without compromising performance.
memory - arxiv:2610.30334 · cs.ROMPBRS: Multi-Potential-Based Reward Shaping for Unified Reward Configuration in Humanoid LocomotionVan Chung Nguyen, Hung Manh La
Designing and tuning reward functions remains one of the main challenges in building effective reinforcement learning (RL) pipelines for humanoid locomotion. While shaping rewards can accelerate learning, naively formulated terms often conflict with the desired behavior, overfit to a specific robot, or become unstable under small weight changes. Potential-Based Reward Shaping (PBRS) offers an alternative solution: shaping the reward via differences of a potential function to guide exploration without changing the optimal policy. In this paper, we study Multi-PBRS for humanoid locomotion and propose a unified reward configuration. We then evaluate the same reward structure and weights across nine humanoid robots, requiring only robot-specific metadata (joint/link mappings and default poses) rather than per-robot reward redesign. Our experiments show that, compared to standard shaping, MPBRS provides more consistent scaling behavior and reduced tuning sensitivity across robots, making it easier to tune without altering the optimal policy.
humanoid - arxiv:2610.30326 · cs.CVR4DGS: Referring Segmentation in 4D Gaussian SplattingBangpu Chen, Yaxuan Li, Shirui Peng, Xiangtian Si +4
Dynamic representations based on 4D Gaussian Splatting (4DGS) have shown strong capabilities in reconstruction and rendering, yet current methods remain limited in supporting complex natural-language referring in dynamic scenes. Existing 4D Gaussian methods are primarily designed for rendering-oriented dynamic modeling, while current semantic extensions often rely on scene-specific semantic optimization or are evaluated on relatively simple query settings. In this paper, we study Referring 4D Gaussian Splatting (R4DGS), a task for grounding temporally varying, multi-target or reasoning-intensive, and zero-target or distractor expressions in dynamic 4D scenes. To support this task, we introduce R4D-Bench-QA, a benchmark with structured query annotations. We further present ReferGaussian, a unified framework built upon 4DGS that converts low-level dynamic Gaussians into persistent, language-addressable entities for training-free referring inference. ReferGaussian organizes mask-supported Gaussian entities in an EntityBank, uses Multimodal Large Language Models (MLLMs) to decompose complex queries into trackable object phrases and structured referring constraints, and performs constraint-aware grounding without scene-specific semantic optimization. Experiments on R4D-Bench-QA and the 4D LangSplat protocol demonstrate improved temporal target selection and competitive spatiotemporal grounding without scene-specific semantic training. Project resources are available at https://chenbangpu.github.io/R4DGS/.
benchmark - arxiv:2610.30320 · cs.ROSelf-Distilling Future Video Representations for World Action ModelsHuiwon Jang, Suhyeok Jang, Hyeonmo Kang, Seungku Kim +2
World Action Models (WAMs) have emerged as a promising approach to learning generalist robot policies using rich spatiotemporal priors from pre-trained video diffusion models. A recent line of WAMs improves inference efficiency by conditioning a separate action decoder on intermediate future video representations extracted from a single denoising step, avoiding costly iterative denoising required for video generation. This design makes action prediction depend directly on the quality of these future representations. In this paper, we introduce Future REpresentation learning via self-Distillation with temporal context (FRED), a framework that trains WAMs to improve the quality of future video representations from noise in a single denoising step. Specifically, FRED uses self-distillation to predict future representations from noise, using representations extracted from ground-truth future observations as targets. We further condition the video model on historical observations, providing temporal context for predicting future representations better. Across diverse real-world manipulation scenarios, FRED consistently outperforms competing WAMs. It achieves particularly large gains on long-horizon and out-of-domain tasks under zero-shot environment generalization, and substantially improves human-to-robot skill transfer, reaching 50.0% task success on tasks absent from robot training data and demonstrated only in human videos, compared with 19.4% for the best competing WAM.
manipulation - arxiv:2610.30319 · cs.ROTerraTransfer++: Demonstration-Free End-to-End PlannerZikang Xiong, Weixin Li, Zhouchonghao Wu, Akshay Rangesh +5
End-to-end autonomous driving has achieved state-of-the-art performance on benchmarks and real-world deployments. Its standard training recipe, however, is expensive across all existing paradigms: imitation learning requires fleet-scale driving logs whose behavior distribution only thinly covers the safety-critical tail; closed-loop reinforcement learning in a photorealistic simulator, while offering diverse behaviors, is bottlenecked by sample-efficiency, per-step cost of rendering, and a forward pass through a large vision backbone. Self-play in vectorized simulators changes the economics: at millions of rollout steps per second, one can obtain a state distribution with natural diversity in collisions, near-misses, and recoveries that no driving log contains, and a highly reactive policy emerges from a basic driving-quality reward. Our approach exploits this asymmetry. We pretrain a single policy by self-play, distill its rollouts into a flow-matching trajectory teacher that also plans, and train lightweight adapters on a frozen pretrained vision backbone to reproduce the latent space of either teacher through the action KL divergence, a batch-relational low-rank structural loss, and, for the planner, a trajectory-matching loss. Every target traces back to the self-play policy, so alignment never supervises against a logged trajectory: a paired dataset of (image, scene-state) frames suffices, with no need for the curated expert demonstrations that imitation pretraining is built on. From this one self-play teacher we obtain two end-to-end students: a closed-loop policy that outperforms every published imitation-based end-to-end method on the photorealistic 3D Gaussian splatting closed-loop benchmark HUGSim, and a planner that reaches 59.4 EPDMS on NAVSIM navhard, above every published method including DriveZero-Scale (57.1), and 95.5 PDMS on navtest, without human demonstrations.
self-playbenchmark - arxiv:2610.30292 · cs.LGGAV: A scientific experimentation environment for PDE agentsMuheng Li, Wenlong Mou
Reliable feedback is essential for coding agents to develop programs for partial differential equations (PDEs). Yet evaluating these programs on new PDE problems often requires expensive reference computations, while fixed datasets restrict the range of experiments. We introduce Generator-Agent-Verifier (GAV), a scientific experimentation environment that makes new PDE problems verifiable by construction. The Generator learns a distribution of continuous reference solutions using data from numerical simulations or experimental measurements to represent a target physical regime. Using the method of manufactured solutions, it works backward from the generated reference solutions to construct matching PDE problems without a separate numerical solve. The Agent designs experiments, investigates failures, and iteratively improves its program. The Verifier evaluates candidate programs against the reference solutions and provides numerical feedback on accuracy and computational cost. This feedback can also guide the generation of new problems that probe the current program's weaknesses. We provide a theoretical analysis of evaluation reliability and guided development. Experiments on Navier-Stokes and Vlasov-Poisson-type equations across different data sources and regimes demonstrate the effectiveness of GAV. In particular, on the numerical Navier-Stokes task, GAV reduces the mean test error of three existing PDE agents by more than an order of magnitude relative to their original setups. Code is available at https://github.com/PotatoJnny/PDE-GAV.
agent - arxiv:2610.30277 · cs.AIBreaking the high-degree barrier in Lehmer's problem with a human-AI discovery loopYuri Chervonyi, Sergei Gukov, Thang Luong
While artificial intelligence has made rapid progress in mathematical reasoning, many open problems depend on structure that is revealed only through large-scale computational exploration and analytical synthesis. One such example is Lehmer's problem from 1933. It asks whether the Mahler measure of an integer polynomial, which measures how far its roots stray from the unit circle, can be made arbitrarily small. Previously, constructing irreducible polynomials with integer coefficients and small Mahler measure has been carried out systematically only for even degrees $d \le 180$. At higher degrees, Boyd's analytical one-parameter family frequently factors, skipping nearly 50% of target degrees and leaving no general construction for arbitrary $d$. Here, we address the high-degree regime through a human-AI discovery loop, in which we extend the conversational prompting with a fast numerical pipeline, an exact symbolic verifier, and large-scale distributed search. The AI agent prototypes numerical and symbolic solvers locally, launches parallel searches across distributed clusters and works with the human researcher to extract and verify mathematical hypotheses. Through this interactive loop, we derive a five-parameter construction, which we call the Generalized Boyd algorithm. It eliminates degree-skipping and constructs irreducible polynomials with Mahler measure strictly below Boyd's limit point $α_1 \approx 1.255434$ for every target degree. Evaluated across all even degrees up to 2000, our approach improves upon six previously published values and holds 99.6% of all best known records. With this work we release the verified dataset of polynomials with minimal Mahler measure across all studied even degrees.
agentai agent - arxiv:2610.30231 · cs.LGEdit, Don't Imitate: Minimal-Edit On-Policy Distillation for Multimodal ReasoningWen Ye, Ailin Deng, Zirui Wang, Shixing Yu +7
Visual mathematical reasoning can fail before reasoning begins: when a vision-language model (VLM) misreads a diagram, subsequent reasoning proceeds from false premises. This perceptual bottleneck is especially pronounced in smaller models, while accurate textual descriptions can substantially narrow the gap to larger models. Because obtaining such descriptions at inference requires additional annotation or model calls, we propose minimal-edit corrective on-policy distillation. Using reference figure descriptions and verified reference solutions as training-only information, a teacher scores incorrect student rollouts under a minimal-edit objective, and only teacher-disfavored tokens contribute to the corrective loss. Correct rollouts receive task-reward reinforcement learning. This preserves on-policy supervision while avoiding full-response imitation. Across model scales, the method requires no descriptions at test time and meets or exceeds the corresponding base model supplied with a reference figure description at inference. It also improves perception-oriented performance, particularly low-level geometric perception, and transfers to external mathematics and perception benchmarks. Because the minimal-edit teacher evaluates corrections rather than generating its own solution, the method remains effective with a frozen same-size teacher, although its gains are smaller than with a larger teacher on most metrics.
benchmark - arxiv:2610.30228 · cs.LGBenchmarking Model Compression Techniques Across Heterogeneous Edge Platforms: A Predictive Maintenance Use Case for Rotating MachineryMatheus Calheira Guimarães de Oliveira, Marco Aurélio da Silva Cruz, Regis Cardoso, Lucas Cambuim +8
To enable Industry 5.0, where autonomous systems are widely adopted and integrated with people to promote sustainable development in industries, it is imperative to adapt complex deep learning models to operate efficiently on resource-constrained hardware for edge inference. However, adapting these models remains challenging due to the hardware heterogeneity of edge computing in industrial settings, performance degradation caused by model compression, and the lack of clear guidance on which compression techniques are practical and effective for applications such as predictive maintenance. To address these challenges, this paper presents a comprehensive comparative analysis of four state-of-the-art model compression techniques applied to a high-performance convolutional neural network for predictive maintenance using a real-world dataset of rotating machinery, equipment typically used in industry. The compressed models were evaluated across a heterogeneous set of nine computing platforms, ranging from highly constrained edge computing devices to general purpose and high-performance processors, reflecting the diversity and fragmentation typical of industrial edge environments. By jointly prioritizing classification performance, inference speed, and compression ratio, we found that compression using knowledge distillation emerged as the most effective approach. The model had more than 90% speedup across all devices while maintaining a 91.22% balanced F-score, with a very small drop in predictive performance relative to the uncompressed model, along with a 98.9% size reduction. Through successful deployment across multiple edge platforms and a systematic comparison of compression strategies, we demonstrated a clear path to achieving Industry 5.0 targets, drastically cutting computational costs and, consequently, energy consumption without sacrificing predictive performance.
benchmark - arxiv:2610.30221 · cs.MAAgentic Architecture of Information Systems. The AAM/5C Model: A Reference Model for Distributed AutonomyLaurent Goujon
Large organizations increasingly face what this book calls the cognitive complexity wall: at the scale of hundreds of interacting components, maintaining the dependency map and reacting at the pace of incidents exceeds what a human team can deliver. This conceptual volume proposes AAM/5C, in which five capabilities (Capture, Comprehend, Choose, Coordinate, Control) are instantiated across seven logical layers, L1 to L7. A single authorization rule, R-AUT, separates who proposes, who decides and who executes: a probabilistic planner (L4) proposes; a governance layer (L6) decides under signed policies and records its verdict in a signed evaluation proof; a security layer (L5) issues a short-lived act token bound to that proof; a policy enforcement point verifies both before the target layer materializes the act. Under an explicit guarantee contract, two analytical statements (H7a, H8a) hold by construction. The other statements of its ten hypotheses (H1 to H10), including log completeness (H7b) and non-bypass of the enforcement point (H8b), are refutable, and the protocol that could refute them is fixed before any measurement. The contribution of the book is a prescriptive reference model that articulates, under this single authorization rule, known mechanisms and adds three constructs, all to be validated: the Validation step of the Distributed Autonomy Loop, the ephemeral authority surface and the cognitive dependency plane. The book positions them against autonomic computing, multi-agent systems, enterprise architecture and agent protocols (MCP, A2A), gives an authorization policy in Rego and a partial TLA+ specification of peer consensus, and sets out a validation program with quantified milestones. It reports no empirical results. This English edition is a translation of the French original, produced with the assistance of AI tools and validated by the author.
agentmulti-agentagenticagent system - arxiv:2610.30206 · cs.LGMoEZip: Routing-Aware KV Cache Compression for Sparse Mixture-of-Experts LLMsMinsang Kim, Seungjun Baek
Mixture-of-Experts (MoE) language models scale model capacity efficiently by activating a small subset of parameters per token. However, in long-context inference, memory remains a major bottleneck because the KV cache grows linearly with sequence length. Existing KV cache compression methods reduce this cost for dense transformers, but overlook a problem specific to MoE: KV eviction can change which experts are selected during answer generation, degrading the generation quality. We propose MoEZip, a routing-aware KV cache compression for sparse MoE language models. To estimate the sensitivity of expert routing to perturbation, we propose a novel metric based on the Fisher information matrix of routing distributions. MoEZip combines this metric with attention weights for routing-aware scoring of contexts. Finally, we compose a complementary scoring system that combines eviction methods for dense and sparse-MoE architectures. Experiments show that MoEZip achieves robust performance under aggressive cache compression on various benchmarks. At a KV retention ratio of 10\%, MoEZip improves over the SoTA baseline by +24 and +20 points on average on LongBench and RULER, corresponding to 2.3$\times$ and 4.9$\times$ higher scores, respectively.
memorylong-contextbenchmark - arxiv:2610.30205 · cs.AIFeedback That Backfires in Games: Do LLM Players Repeat the Move They Just Watched Fail?Esmail Gumaan
When an LLM agent's move fails, the harness writes the move and the failure message into the transcript and asks again. In tool calling, this record makes small models more likely to repeat the failed action. We test whether the same holds in games (Sokoban, Connect Four, Othello), where a move can fail by being illegal or by being legal but losing, both verified exactly. On fixed probe items built by perturbing a correct move, the record backfires: pooled over 8 instruction-tuned models (135M-1.7B) it makes the failed move more likely (corrective gain G = -1.38 nats per token), through the move's surface form, for illegal moves but not for losing ones, and not for every model. An exploratory analysis shows what sets the sign: after the record, the failed move's log-probability barely depends on how likely it was before (slope 0.18). Moves the model would not have played become more likely, and greedy decoding starts to produce them; moves it was about to play become less likely, but greedy decoding still repeats 38% of them. A description of the failure that omits the move string keeps the model's preference instead. Along the OLMo 2 1B post-training ladder the inversion is present in the base model and largely removed at the DPO stage. RAGEN's default models invert in RAGEN's own Sokoban format before any RL. A "do not repeat" instruction does nothing, a clean retry is the worst harness, and a decoder ban stops exact repeats but not illegal play. All measurements ran in float32 on a free Colab GPU, which first reproduced the original paper's number.
llm agenttool callingpost-training - arxiv:2610.30202 · cs.AIMPIDetector: Detecting Multi-modal Prompt Injection via LLM-based ReasoningYiheng Li, Jingyue Li
With the rapid integration of Multi-modal Large Language Models (MLLMs) into real-world applications, Prompt Injection (PI) has emerged as a critical security threat. Attackers can conceal malicious instructions within textual narratives or visual inputs to bypass safety alignment. Existing approaches, such as signature- and perplexity-based methods, rely heavily on pattern matching, making them insufficient to detect sophisticated, cross-modal PI payloads. In this paper, we propose a novel detection framework, called MPIDetector, that leverages LLMs' intrinsic reasoning capabilities via few-shot In-Context Learning (ICL). Instead of fine-tuning, we provide the model with specific prompt templates and representative attack examples, enabling it to scrutinize the semantic intent of input across modalities. We constructed a comprehensive evaluation benchmark comprising eight textual and seven visual malicious categories, supplemented with benign datasets to rigorously test for false positives. Our evaluation across multiple leading LLMs, including Gemini-2.0-Flash, Qwen3-VL-32B-Instruct, and Doubao-seed-1.6-flash, demonstrates the efficacy and generalizability of this approach. The proposed method achieves up to 90.91% accuracy and 100% recall on textual tasks. In the visual domain, it achieves up to 99.28% accuracy while maintaining 100% precision across all evaluated models. Crucially, when evaluated against 500 highly obfuscated environmental web-agent injections (WebInject), MPIDetector achieved a detection rate up to 99.80%. Comparisons with two baseline approaches for detecting textual PI payloads, along with two additional methods for visual input detection, demonstrate MPIDetector's superior accuracy. These results show that LLM-driven reasoning provides a robust, cross-modal defense against advanced prompt injections.
benchmark - arxiv:2610.30200 · cs.AIDetection Is Not Enough: Reasoning Traces as Testable Program HypothesesVaidehi Hegde
When a language model is deployed, no reference solution exists, so the reasoning trace it writes is the only evidence for judging its answer. The reference-free substitutes each fail differently: self-assessment degrades the answer it was meant to protect, sampling agreement measures stability rather than correctness, and perturbation studies read the expected answer off a human-written template, grading a model rather than a trace. We ask how far a single trace can be verified with no reference solution, no judge model and no weight update, and find, for a frozen LLaMA-2-7B-chat on GSM8K, that detection is not what limits such a pipeline. Checking a trace against itself reaches an AUROC of 0.741 and stops there for a structural reason: two thirds of wrong traces state no false equation, having executed a mistaken plan correctly, and on wrong-plan failures the detector flags a wrong trace less often than a right one. A candidate-derived metamorphic probe exposes many of these failures: we abstract the trace into a program over the question's quantities, perturb one quantity, execute the program, and ask the model the rewritten question once, so every trace supplies its own test. Valid perturbations exist for 937 of 1,319 test questions (71.0%); among the 847 that also have a parseable first trace, supported traces are correct 60.6% of the time against a 31.6% base rate, although the probe also refutes 45.5% of correct traces. No tested policy yields a statistically significant improvement in accuracy. A flag earns its retry only when detector precision exceeds D/(R+D), where R and D are the rates at which a retry repairs a wrong trace and damages a correct one; our repair rate is so low that, at the same retry budget, even a perfect detector would recover an estimated 66 of 1,319 questions. Repair, not detection, is the open problem.
judge model - arxiv:2610.30197 · cs.ROAlpha: Systematic Fault Injection and Robustness Evaluation for LLM-Driven Embodied ManipulationJona Joy
Large language models (LLMs) are increasingly integrated into embodied robotic planning, yet standard benchmarks primarily evaluate end-to-end task completion in nominal environments, obscuring failure origins across the execution stack. We present Alpha, a systematic fault-injection and robustness-evaluation framework for LLM-driven manipulation in simulation. Alpha injects calibrated faults across perception (sensor lag), actuation (grip slip), kinematics (unreachable targets), and planning (planner-output corruption), alongside concurrent conditions. To separate physical vulnerabilities from planner sensitivities, we introduce a matched methodology pairing an LLM agent (Llama-3 via Ollama) against a deterministic Scripted Oracle planner across a balanced 150-episode matrix under identical seeds, geometries, and safety criteria. Across the matrix, the Oracle achieves 64.0% (48/75) success while the LLM achieves 44.0% (33/75) (paired McNemar p = 3.01 x 10^-4). Crucially, excluding planner-output corruption, performance across all physical and kinematic conditions is identical: 55.0% (33/60) for both planners with zero discordant pairs (McNemar p = 1.00). This demonstrates that physical actuator slip, kinematic reach boundaries, and perceptual staleness affect deterministic controllers and LLM planners equivalently. The performance gap is driven entirely by planner-output corruption (100% Oracle vs. 0% LLM), isolating the prompt-to-JSON serialization and validation interface rather than a cognitive reasoning deficit. We also audit 169 historical pilot attempts for transparency.
embodiedmanipulationagentllm agentbenchmarkevaluation framework - arxiv:2610.30193 · cs.CLProtection Is Not One Event: A Survey of RAG Protection Research by Measured EndpointPeiyang Liu, Xi Wang, Di Liang, Youru Li +2
Retrieval-augmented generation (RAG) systems face security, privacy, and copyright risks including content leakage, source reconstruction, unauthorized access, data poisoning, and prompt injection. Yet RAG protection studies use incompatible evaluation metrics, datasets, and threat models. We present a systematic survey and human-coded evidence map of 118 full-text-confirmed RAG defense papers, organized by measured endpoint rather than claimed goal or defense mechanism. Our seven-family taxonomy covers content disclosure, formal privacy proxies, access control and suppression, corpus-use detection, and pipeline integrity and attribution. Thirty-two annotators double-coded the papers kappa=0.867. The survey identifies three gaps. First, retrieval reranking and context rewriting each span five or six endpoint families, so mechanisms do not identify protection outcomes. Second, only 17.8% of studies use content disclosure or reconstruction as their primary endpoint; under the frozen copyright/ownership relation label, 73.3% of records have corpus-use detection as their primary endpoint, and none have an emitted-expression primary endpoint. Third, most evaluations use general-purpose question-answering datasets and non-adaptive attacks rather than sensitive data and adaptive, multi-turn extraction. Our controlled RAG testbed shows that preventing verbatim copying does not prevent private-fact recovery: a normalized 8-gram guard reduces 20-token copying to 0.0%, but a three-query adaptive attacker recovers 48.3% of hidden facts on Qwen3-8B. We release the taxonomy, dataset, coding protocols, and testbed to enable comparable, claim-aligned RAG security evaluation.
retrieval-augmentedrag - arxiv:2610.30188 · cs.LGAgentic Verified Optimization of Real-World SoftwareZhe Ye, Udaya Ghai, Zhizhen Qin, Dawn Song +1
Program optimization aims to improve performance while preserving behavior. Coding agents can already make programs faster by changing algorithms, data structures, and execution strategies. However, existing equivalence checks mostly rely on tests, which agents have been shown to exploit, passing the tests without preserving the intended behavior. We propose agentic verified optimization, in which an agent produces a faster implementation together with a machine-checked proof of equivalence to the existing one. Because the existing implementation serves as the specification, the task avoids the specification-writing bottleneck of verified code generation, and recent advances in theorem proving make the proofs attainable. We formulate the task at the function and repository levels in Lean 4, where programs can be both executed and reasoned about. Few real-world applications are written in Lean, so we develop Aqueduct, an agentic framework that migrates C, C++, Go, and Rust programs into native Lean, validates each port by differential testing against its original. On the 189 reference implementations in Verina, agents produce faster proved rewrites for 156, including algorithmic changes such as replacing exponential recursion with dynamic programming. On 17 migrated applications, their 92 proved rewrites speed up 47 of 51 application scenarios. These application-level gains are smaller than the function-level ones, making the choice of functions to optimize an important problem. Current agents can thus already carry out verified optimization on both standalone functions and applications migrated from real programs. We discuss the remaining challenges and the steps toward verified optimization in production.
agentagentic - arxiv:2610.30184 · cs.AISelf-Evolving Static Analysis via LLM Pseudo-LabelingRui Melo, Andre Duarte, Aditya Oke, Shubham Gandhi +4
Security bugs often slip past static analysis not because they are impossible to detect, but because detection rules have not yet been written for the code in front of us. New frameworks, APIs, and coding idioms continually create such gaps, leaving developers to discover failures and manually encode new rules. We present SESAA, a framework in which an Large Language Model (LLM) agent starts from no rules and iteratively evolves a set of readable and auditable joern queries. The agent receives feedback only from a second LLM critic, which inspects flagged and unflagged code to identify likely false alarms and suspected misses, without access to ground-truth labels. The agent then uses this feedback to revise its queries. On the OWASP Benchmark for Python, SESAA reaches a mean F1 of 0.742 across 8 independent runs, well above the default configurations of five widely used tools and comparable to the strongest of them once it is tuned with access to ground-truth labels that SESAA never sees. In a pilot on real-world Python projects, queries induced from code available before a vulnerability was introduced detect 10 of 19 subsequent CVEs, including 9 detected by rules written for the same repository. The resulting detector is a set of executable, human-readable queries that engineers can inspect, modify, and run independently. These results suggest that LLM agents can turn static-analysis rule maintenance from a manually authored process into an iterative process of automated discovery and repair.
agentllm agentself-evolvingbenchmark - arxiv:2610.30179 · cs.AIGeoDiv KAN Solver for Time Fractional Navier Stokes Equations on Complex Domains to Realistic CFD ReconstructionYatin Kumar, Shivani Saini, Ramesh Kumar Vats
Scientific machine learning provides a promising framework for solving complex problems in science and engineering. Notably, physics informed neural networks (PINNs) have shown considerable potential for solving differential equations and modelling engineering systems. Solving time fractional Navier Stokes equations (TFNSEs) on complex domains is challenging due to temporal nonlocality, nonlinear flow dynamics, geometric complexity, and incompressibility constraints. To address these challenges, we propose a geometry gated, divergence-free Gaussian radial basis function Kolmogorov Arnold network (GeoDiv KAN). The proposed architecture uses geometry dependent feature gating and a streamfunction formulation to preserve incompressibility by construction. GeoDiv KAN is evaluated for the time fractional incompressible Navier-Stokes equations on square, circular, and star-shaped domains. Numerical experiments show that GeoDiv KAN achieves higher predictive accuracy than PINN, PIKAN, RBF KAN, and DF RBFKAN across the considered benchmark problems. To establish practical relevance, GeoDiv KAN is applied to two data driven CFD cases: unsteady flow past a circular cylinder using the \href{https://doi.org/10.5281/zenodo.5039610}{Zenodo dataset} and turbulent RANS flow through a complex U bend using the \href{https://huggingface.co/datasets/JDecke/ubend-cfd}{U-bend dataset}.
benchmark
02 US SEMI · SEC 8-K FILINGS
2 itemsscanned: NVDA / AVGO / MRVL / COHR / LITE / AMD / TSM / SMCI / ANET / CRDO / POWL / VECO
03 HUMANOID · COMPANY NEWS
62 itemsscanned: figure-ai / 1x / boston-dynamics / unitree / apptronik / sanctuary-ai / neura-robotics / agility-robotics / physical-intelligence / agibot
Figure AI (10)
- Figure AISeptember 30, 2026F.02 Decommission
- Figure AISeptember 17, 2026Helix 2.5: Zero-Shot 30-Home Generalization
- Figure AIOctober 09, 2025Introducing Figure 03
- Figure AISeptember 03, 2026Figure and Nscale Sign Strategic Partnership For Up to 100,000 GPUs on the NVIDIA Vera Rubin Platform
- Figure AIAugust 25, 2026Introducing Index: Building The World’s Largest and Most Diverse Physical Dataset
Boston Dynamics (10)
Unitree 宇树 (10)
- Unitree 宇树Components
- Unitree 宇树Kung Fu Meets Spring, Unitree SFG Robots Present "Cyber Real Kung Fu" in the Year of the Horse2026-05-31Media Coverage
- Unitree 宇树Welcoming Myanmar President Min Aung Hlaing to Unitree2026-08-05Media Coverage
- Unitree 宇树Unitree founder Wang Xingxing graces the cover of Time magazine2026-08-05Media Coverage
- Unitree 宇树Unitree Announces H2 Plus, an NVIDIA Isaac GR00T Reference Humanoid Robot for Academic Research2026-06-01Media Coverage
Apptronik (1)
Sanctuary AI (6)
- Sanctuary AIBlog PostTIGER: Task-Space Imitation Guidance for Efficient Reinforcement LearningDemonstrations as a compass, not a scriptRead More
- Sanctuary AIPress ReleaseProduct UpdatesSanctuary AI Expands Physical AI Strategy to Industrial Robotics, Demonstrating Production-Ready AI PerformanceRead More
- Sanctuary AICorporate NewsDaniel Friedmann Appointed CEO of Sanctuary AIRead More
- Sanctuary AIPress ReleaseZeon Invests in Sanctuary AI and Partners to Advance Specialized Materials for Dexterous RoboticsRead More
- Sanctuary AIThought LeadershipWeb Summit Reflections: Canada’s Physical AI Moment Can’t WaitRead More
Agility Robotics (10)
- Agility RoboticsAgility’s Humanoid Deployment ProcessPras VelagapudiAugust 04, 2026
- Agility RoboticsJonathan Hurst answers your questions about building Digit 5InsightOctober 23, 2026
- Agility RoboticsThe Realistic Pathway to HomeInsightMay 26, 2026
- Agility RoboticsAgility and AIInsightMarch 16, 2026
- Agility RoboticsAgility Gets a New BrandInsightMarch 5, 2026
Physical Intelligence (8)
- Physical IntelligenceNewFrontier Models for Every RobotOctober 13, 2026Our latest robotics models are now available through the Pi API beta. Upload data, fine-tune models, and deploy cloud-hosted inference on your own robots.
- Physical Intelligenceπ0.7: a Steerable Model with Emergent CapabilitiesApril 16, 2026A steerable robotic foundation model that exhibits a step-change in generalization.
- Physical IntelligenceThe Physical Intelligence LayerFebruary 24, 2026General-purpose physical intelligence models will enable a Cambrian explosion of robotics applications. See how our partners are already solving real-world problems.
- Physical IntelligenceMoravec's Paradox and the Robot OlympicsDecember 22, 2025By fine-tuning our latest model, we were able to solve a series of very difficult manipulation challenge tasks.
- Physical Intelligenceπ*0.6: a VLA that Learns from ExperienceNovember 17, 2025A method for training our generalist policies with RL to improve success rate and throughput on real-world tasks.
智元 AgiBot (7)
- 智元 AgiBotAGIBOT Open-Sources AGIBOT WORLD Release 4: "Physical Reasoning" Dataset Enabling Robots to Reason Before Acting2026-10-23
- 智元 AgiBotAGIBOT Opens Hong Kong Office in Causeway Bay2026-10-09
- 智元 AgiBotAGIBOT Partners with Vietnam National Innovation Center(NIC) to Advance Embodied AI Industry and Talent Ecosystem2026-10-02
- 智元 AgiBotIDC Data Shows AGIBOT Led Global Humanoid Robot Shipments in H1 20262026-09-30
- 智元 AgiBotAGIBOT and ASD Deploy 100 AGIBOT X2 Humanoid Robots Across 100 Retail Stores2026-09-29