Physical AI Brief

Multimodal large language models (MLLMs) can interpret a street view, but urban agency depends on whether such local evidence remains useful after the agent starts to move. In this paper, we investigate how far current MLLM agents can turn local urban perception into reliable action in a complicated real-scale city. We propose UrbanGround, the first sandbox to make this question testable in a physically constrained replica of Hong Kong built from territory-wide 3D geospatial data. UrbanGround supports closed-loop interaction from a first-person view and provides an interactive map for navigation. Agents can directly enter the 3D city and explore from a first-person view. Our analysis follows the growth of the spatial problem through three research questions. We first test whether an agent can ground a local scene well enough to answer spatial questions after active observation. Then we ask whether that grounding supports navigation as destinations become farther away and less explicit. Finally, we examine whether the resulting behavior survives changes in route availability and pedestrian motion. Contemporary MLLM agents usually show useful atomic abilities in visual recognition and short-range spatial reasoning, while orientation and pedestrian-aware movement remain unreliable. Their central failure emerges over extended exploration, where local abilities do not compose into sustained goal-directed behavior and errors accumulate without effective correction. We hope UrbanGround will support broader study of how far current MLLM agents can explore reliably in complex, open-ended urban environments.

Agent skills package specialized knowledge and workflows into reusable resources that extend AI agent capabilities. Recent work automatically discovers such skills from agent experience, which enables agents to progressively adapt through interaction. However, the insights that guide skill development typically remain scattered across optimization histories, limiting their systematic reuse across iterations. We introduce WikiSkill, a framework that co-evolves agent skills with a persistent knowledge base (wiki). At a high level, WikiSkill separates raw execution experience, accumulated knowledge, and executable skills, while continuously consolidating experience into the wiki, which subsequent skill updates can build on. Across diverse benchmarks and models, WikiSkill consistently outperforms state-of-the-art skill-evolution methods and improves over no-skill baselines in most model-benchmark settings. We find that skill evolution complements model scaling: larger models generally benefit more from evolved skills, while smaller models with skills can outperform substantially larger models without them. We also find that evolved skills transfer effectively across models and model families, and skills evolved by other models can outperform self-evolved skills. Finally, our ablation studies confirm that persistent knowledge accumulation in the wiki is critical for effective skill evolution. These results demonstrate the benefits of systematically accumulating and refining agent experience for developing reusable and transferable skills.

To improve large language models' ability to resolve real-world software issues, prior work has focused on constructing large-scale agent trajectory datasets and performing supervised fine-tuning (SFT) on successful trajectories. However, task success does not guarantee high-quality supervision: successful trajectories may still contain ineffective, redundant, or risky steps. Directly using such trajectories for SFT can introduce noisy supervision and encourage models to imitate undesirable problem-solving behaviors. Therefore, we propose SWE-Prime, a multi-granularity, two-stage SFT data selection method that progressively filters training data at the trajectory and segment levels. Specifically, the first stage performs trajectory-level screening based on process quality, result quality, and data representativeness, selecting a high-quality and representative subset of successful trajectories. The second stage performs segment-level selection by grouping consecutive steps into semantic segments and assessing each segment based on its contribution to the final solution, learnability, and potential risks. During SFT, all segments remain in the sequence to preserve context, while only selected segments contribute to the loss computation. Experiments on SWE-Bench Pro and SWE-Bench Verified show that training on the 10% trajectory subset selected by SWE-Prime outperforms training on the full resolved dataset, yielding relative performance gains of up to 12.2% and 24.2%, respectively.

Recent prominent post-training methods, such as Reinforcement Learning (RL) and On-Policy Self-Distillation (OPSD), have driven rapid progress in mathematical reasoning for large language models, yet their reliance on ground-truth labels precludes test-time training (TTT). Replacing ground truth with majority-vote pseudo-labels is a natural alternative, yet it is fragile: an incorrect vote corrupts the teacher and misleads every token. We observe that this failure mode is asymmetric: rollouts that disagree with the pseudo-label are typically wrong regardless of whether the vote itself is correct. Building on this observation, we propose Test-Time Policy Optimization (TTPO), an asymmetric objective that distills agreeing rollouts via OPSD and penalizes disagreeing rollouts with Grouped RL. Token-level selection further refines both branches: distillation down-weights already-converged positions, while RL penalizes only confident errors. Both updates remain well-grounded even under frequent pseudo-label errors, and majority-vote routing yields tighter self-supervision as the model improves. Without any labels, TTPO matches label-supervised OPSD on five competition-level benchmarks, raises Qwen3-1.7B from 38.0% to 45.2% in TTT, yields +25.2% to +36.4% without thinking, and shows strong cross-task generalization.

In real-world software development, code review typically involves iterative interactions between developers and reviewers to improve software quality, making the process costly and time-consuming. Although recent work explores large language models (LLMs) for automated code review, most approaches oversimplify code review into a single-round, static decision task, which fails to capture the multi-round interactive nature and the complex problem-solving processes inherent in realistic review scenarios. To bridge this gap, we introduce MCR-Bench, the first defect state-aware benchmark designed for realistic multi-round code review. MCR-Bench covers five commonly-used programming languages and consists of 2,269 real-world multi-round code review tasks, each of which is annotated with fine-grained defect information and cross-round state labels. Each task in MCR-Bench is equipped with fine-grained defect metadata (e.g., description, type, severity) alongside dynamic state annotations, capturing the complete evolutionary trajectory of a defect throughout the multi-round process. We obtain several findings through extensive experiments on MCR-Bench with mainstream LLMs. (1) Limited overall capability: experiments reveal that mainstream LLMs exhibit limited overall performance in defect detection and defect lifecycle state tracking, with performance degrading significantly as the number of interaction rounds increases; (2) Defect-sensitive performance: LLMs' performance varies substantially across different defect types and severity levels, with semantically complex or low-salience defects being significantly more likely to be missed; (3) Underlying Failure Mechanisms: our in-depth error analysis dissects the distinct drivers of false positives and false negatives, revealing critical weaknesses such as cross-round temporal misalignment and inadequate long-range memory.

LLM-based agents are increasingly deployed in product-level execution harnesses, where jailbreaks can trigger harmful tool use and persistent state changes, creating greater risks than unsafe text generation alone. Existing automatic red-teaming methods often rely on fixed attacks, while recent agentic attackers coordinate multiple jailbreak tools and show stronger potential through trajectory-based retrieval. However, such retrieval can reuse misleading experiences due to retrieval bias and unclear tool credit, and full trajectories add context overhead while reducing interpretability. We propose RedEvoAgent, a black-box red-teaming agent that distills cross-case attack trajectories into a concise, human-readable attack skill. The attack skill adaptively evolves through tool-effectiveness profiling and Deciding-Tool Attribution for skill updates, and a validation ratchet that retains only updates improving validation performance. Experiments on multiple benchmarks, target models, and target execution harnesses show that RedEvoAgent outperforms fixed and agentic baselines, improves tool efficiency, and transfers across attacker models and target execution harnesses.

Chemical reactions are fundamentally transformations in electron space, yet most machine learning approaches model them either through \textit{de novo} generation of product molecules or through heuristic graph edits that operate directly on molecular topology. We introduce MAELLE (\textbf{M}ech\textbf{A}nistic \textbf{E}dit f\textbf{L}ow-matching on e\textbf{L}ectron r\textbf{E}arrangements), which instead models reactions as discrete flow matching over electron occupation vectors. Concretely, we formulate the reactant-to-product mapping as a Continuous-time Markov Chain (CTMC) over the graph-structured integer-valued electron occupation space defined on all bonding, non-bonding, and hydrogen sites. To construct the intermediate edit trajectories, we generalize the discrete flow matching mixture path to discrete electron rearrangements using Optimal Transport, yielding a sequence of mechanistically interpretable edit moves without requiring elementary step annotations. MAELLE achieves competitive performance on the USPTO-480K benchmark compared with leading reaction prediction models. Beyond in-distribution accuracy, we evaluate robustness across two out-of-distribution settings - structural complexity and reaction type - and find that MAELLE maintains strong performance where existing methods degrade. Finally, because the learned flow operates over the full electron redistribution, MAELLE naturally recovers mechanistic trajectories that align with known chemistry and can predict side products of a reaction.

Large language model (LLM) agents in governed organizations must let the persona (instructions, tone, self-presentation) evolve freely, while keeping execution (stateful, audited work) traceable. A single trust domain does not satisfy both cheaply. We present Persona-Execution Separation (PES): persona and execution reside in different trust domains, connected by a governed contract bridge. The persona is singly-homed and may drift; execution is faceless and audited. Status summaries may return; data bodies remain in the restrictive domain except a graded data-loss-prevention (DLP) exception; identity stays continuous. An approval matrix, DLP, and audit enforce the crossing. PES follows from three goals---free drift, execution traceability, and decoupling. Under LLM representational indistinguishability, any single-domain mechanism that meets all three must re-introduce typed change objects, an external gate, and a stable audit anchor: PES rebuilt at higher coupling cost. A development/pilot case in a regulated digital-employee platform records five decisions over one month, each with a rejected alternative. A mechanism check on the shipped implementation found no execution-side re-validation under persona perturbation (five model configurations) and no persona fingerprint on hard-asserted fields. A probe of a recovered pre-separation build found the governed execution path decoupled from the persona by omission, not by construction; a later wiring change could reverse that isolation, which PES makes an audited architectural rule. The pattern applies when multi-user deployment, execution audit, and expected persona churn hold jointly.

Static scanners are increasingly used to identify executable or otherwise unsafe content in machine- learning artifacts, yet conventional evaluation metrics characterize only cases where a scanner yields a usable security judgment. We evaluate ModelScan, ModelAudit, and Fickling using a controlled, artifact-backed benchmark on a synthetic corpus of 170 Pickle and PyTorch focused artifacts across 145 specimen families, 135 of which have binary security ground truth and 10 of which are intentionally malformed without labels. We explicitly distinguish non-N/A coverage, analysis completion, definitive security decisions, non-security findings, and unsupported outcomes. On labeled families, ModelAudit produced definitive security decisions for all 135 families (100%), Fickling for 110 (81.5%), and ModelScan for 67 (49.6%). Conditional on making a definitive judgment, ModelScan achieved 100% precision, recall, and F1. Fickling identified no unique true- positive families beyond those found by the combination of ModelAudit and ModelScan. Furthermore, for the 48 malicious families where ModelScan failed to complete its analysis, both ModelAudit and Fickling generated detections consistent with ground truth. These findings underscore the need to separate judgment accuracy from judgment availability, as well as incremental detection coverage from tool-level redundancy.

Reinforcement Learning with Verifiable Rewards (RLVR) significantly improves LLM reasoning but often causes a drop in policy entropy, leading to narrowed reasoning coverage and degraded pass@$k$ for large $k$. While existing methods mitigate this entropy collapse through algorithmic regularizations, cross-model non-parametric perturbation is also neglected. In this work, we propose a simple yet effective approach to preserve the generative diversity of LLMs during RLVR. Instead of relying solely on internal exploration, we force the target model to generate answers based on partial reasoning trajectories generated by a smaller, weaker language models. These unfamiliar prefixes effectively disrupt over-confidence and encourage the exploration of distinct reasoning paths. We empirically study the potential of outer prefixes, revealing the mechanism of the impact of distributional discrepancy to the exploration dynamics in RLVR training. Experiments across multiple mathematical benchmarks show that our method consistently outperforms vanilla RLVR. Notably, the performance gain becomes increasingly pronounced as $k$ scales up, demonstrating a substantial expansion of reasoning coverage. Furthermore, our approach efficiently mitigates entropy collapse without requiring additional SFT, intricate reward designs, or complex prompting.

Vision-language models (VLMs) can locate an image region referred to by a text prompt and route the corresponding visual evidence to the output, yet the internal mechanism behind this behavior is not understood. Inspired by retrieval heads in large language models, we ask whether VLMs contain an analogous mechanism for visual retrieval. We answer affirmatively by introducing Visual Retrieval Heads (VRHs), a small subset of attention heads (about 1.7-2.6%) that are causally responsible for grounding text descriptions to image regions. To find them, we recast existing head-scoring methods under a unified design space over query tokens, key aggregation, and cross-sample aggregation. We then show that scoring attention from output prediction tokens with a sum over the ground-truth referent region most reliably identifies causal heads. Across eleven VLMs and five referring-expression benchmarks, masking only the top 20 VRHs reduces grounding accuracy by up to 80 percentage points, while masking the same number of random heads has little effect. Beyond replicating the causal-sparse-universal triad established for text retrieval heads, VRHs exhibit several properties not previously reported: they generalize across visual reference tasks, remaining causal on attribute, spatial, counting, and visual-math benchmarks despite being discovered through bounding-box prediction; they are functionally specific, preserving output format while corrupting localization; and they are architecturally shared, transferring causally across VLMs that share an LLM backbone but differ in vision encoder, projector, and instruction tuning.

Reinforcement learning with verifiable rewards (RLVR) improves specific capabilities of large language models, but covering multiple capabilities often involves training separate domain experts and subsequently consolidating them. We organize three fusion paradigms by the artefacts they reuse: Merge combines expert task vectors, Mix RL pools their datasets, and multi-teacher on-policy distillation (MOPD) uses both. Because they have largely been studied in isolation, how they compare and how to choose among them remain unclear. We compare all three using shared experts and data across model scales and a multi-domain benchmark suite. Although their average performance differs by at most 1.4 points, the gap reaches 8.6 points on a single benchmark, with domain-level variation tracking cross-domain relations visible in task-vector geometry. Training dynamics expose distinct constraints: Mix RL depends on domain mixture proportions, MOPD remains bounded by its teachers, and Merge compresses all expert updates into one. All three improve single-sample accuracy without measurable gains in solution coverage or losses in held-out capabilities. These results yield a practical guideline: use Merge when experts already exist and cheap fusion is paramount; Mix RL when training a unified model without experts, with domain proportions adjusted for cross-domain transfer; and MOPD when preserving domain-specific gains matters more than surpassing teachers or minimizing end-to-end cost.

Estimation of Human-Object Interactions in 3D (3D HOI) is a fundamental problem in 3D computer vision with applications in AR/VR, robotics, and embodied AI. However, reconstructing these interactions in 3D remains challenging due to depth ambiguities, occlusions, and object shape variability. Existing approaches are primarily concerned with reprojection and contact constraints, fitting parametric human models and object templates to 2D images. In this paper, we explore a different avenue. We present MILO, a framework that leverages the visual capabilities of Large Reconstruction Models (LRMs) to recover detailed 3D human-object interactions from a single image. Our key observation is that LRMs provide a powerful geometric scaffold that preserves relative human-object arrangement and proximity cues. This significantly simplifies the reconstruction procedure, reframing the problem as interpreting the LRM mesh: we segment it into human and object components, fit a parametric body model to the human part, and optionally align an object template to the object part (if such a template is available). MILO achieves strong reconstruction accuracy and outperforms existing baselines across multiple benchmarks and interaction scenarios. Our code is available at https://ac5113.github.io/MILO.

State-of-the-art action-conditioned video models are typically restricted to a single robot embodiment, preventing them from leveraging the vast corpus of heterogeneous video data that contains rich signals for learning generalizable physics. To bridge this gap, we introduce CLAP, a framework for cross-embodiment action-conditioned video generation capable of being trained on diverse, internet-scale videos across human and robotic agents. CLAP is grounded in the insight that universal physical laws govern spatiotemporal dynamics regardless of the actor. However, cross-embodiment learning is non-trivial because action representations vary sharply across robot platforms and are typically absent in human videos. CLAP addresses this fundamental challenge through the following core contributions. First, CLAP reconciles disparate action spaces using end-effector poses, language instructions, and latent actions. Second, to resolve their individual limitations, CLAP introduces a curriculum-based cross-embodiment learning recipe that first learns foundational physical priors across unlabeled video data using latent actions and subsequently grounds them in end-effector action spaces for zero-shot deployment to real-world tasks. Crucially, CLAP approaches or surpasses state-of-the-art single-embodiment video models in challenging environments like DROID. These performance advantages compound via few-shot adaptation to establish a novel paradigm for training single-embodiment video world models. Ultimately, CLAP delivers the most comprehensive suite of action-conditioned video world models to date - spanning diverse action-conditioning spaces (end-effector, language, and latent) and robot morphologies (including cross-embodiment, DROID, Bridge, bimanual YAM robots, and G1 humanoids). We open-source all code and models. Project Website at https://omni-clap.github.io .

Video carries the temporal structure of the physical world, yet learning representations from it has remained computationally expensive: prevailing self-supervised methods either prevent representation collapse through architectural asymmetries, coupling an exponential-moving-average target encoder, a stop-gradient, and a capacity-limited predictor, or circumvent it by reconstructing masked content in pixel space. We introduce LeVJEPA, the first video encoder trained under LeJEPA's collapse-free objective, which dispenses with both. A single encoder is trained with an invariance loss over global and local views of a clip, regularized by SIGReg, which excludes collapse with a provable guarantee. The architecture reduces to an encoder and a projector, and the objective to a single hyperparameter. This formulation admits two properties. First, the cost of pretraining is governed by the number of tokens the encoder observes; uniform random token dropping renders this number small while simultaneously improving downstream accuracy. At matched epochs on identical data, LeVJEPA matches or surpasses V-JEPA 2 across ViT-S/B/L at 5.6 to 20.8x less pretraining compute, and at matched total FLOPs it exceeds the strongest video baseline by 7.6 points on ImageNet-1K while remaining competitive on motion-centric benchmarks. Second, since no asymmetry between branches is required, the encoder can be trained with block-causal attention at no measurable accuracy cost: temporal ordering becomes a property of the encoder itself. Against a compute-matched DINOv2 trained on frames of the same videos, LeVJEPA approaches the image-pretrained encoder on appearance-centric evaluation while nearly doubling its motion-centric accuracy. These results indicate that, once its computational overhead is removed, video becomes a viable and in several respects preferable substrate for general-purpose visual pretraining.

Retrieved scientific literature can serve as inspiration for both human and AI scientists. Inspiration can take different forms: prior work may directly suggest how to address a problem, or surface directions at different levels of abstraction - zooming out to a more general view or zooming in to a concrete realization. We introduce RATIO (Retrieval Across Typed Ideation Operations), a large-scale benchmark in which relevance is defined by three operations which we name ideation moves: Address retrieves potential approaches for stated problems, Broaden retrieves more general formulations, and Specify retrieves concrete instantiations. RATIO is constructed from millions of full-text scientific papers across CS literature via a general recipe that extends discourse-marker distant supervision - previously used only for classification - to corpus-scale retrieval, combined with extensive LLM and human vetting. Experiments show that operation-specific fine-tuning substantially boosts retrievers but leaves much room for further improvements. RATIO provides a scalable training and evaluation framework for retrieval components that support literature-grounded ideation, opening up new research avenues on scientific inspiration retrieval.

LLMs are increasingly able to answer complex questions about enterprise-scale document collections. But evaluation is hard: companies don't want to share internal communications, and synthetic datasets have been overly simple. We present CorporateBench (CB), a human-validated multi-task Q&A benchmark whose scale approaches the conditions LLMs encounter in corporate communication networks, with evaluation corpora surpassing 230,000 documents. CB evaluates LLMs across two dimensions (information extraction and knowledge base querying) through four synthetically generated firms ranging from 12 to 10,000 employees. Each corpus is sampled from a temporally evolving knowledge base describing a consistent world, guaranteeing cross-document logical consistency even across hundreds of thousands of documents. We evaluate five LLMs on CB, revealing increasingly poor performance as input size approaches realistic scales. CB provides LLM developers a metric for corporate communication reasoning, filling a crucial gap in the benchmarking ecosystem.

Vision-Language-Action (VLA) models are increasingly promising for robotic manipulation, yet their real-world deployment remains bottlenecked by high inference latency and unstable asynchronous execution. This challenge is particularly pronounced in flow-matching-based VLA models, where action decoding requires multiple iterative steps conditioned on the VLM context. While efficient inference methods improve control frequency and asynchronous methods reduce execution idle time, existing approaches often fail to jointly achieve low-latency inference and accurate, temporally consistent asynchronous execution. We introduce \textbf{FlashVLA}, a streaming action decoding framework that addresses both challenges in a unified formulation. FlashVLA maintains a streaming action buffer with multiple chunks at different noise levels and decodes them using chunk-wise causal attention. This design allows FlashVLA to produce one executable action chunk per inference step. Moreover, its chunk-wise autoregressive formulation implicitly preserves action continuity, enabling smooth asynchronous execution without extra future-state conditioning. Across extensive simulated and real-world experiments, FlashVLA substantially improves inference speed while maintaining strong task performance. It can achieve $\geq$30\,Hz control frequency on a single GPU with smooth asynchronous inference in real-world deployment.

AI-generated text detection is commonly framed as a binary document-level judgment about whether a text is human-written or machine-generated. This framing breaks down for mixed-origin writing, where content origin and expression origin may differ. We cast mixed-origin detection as dimension-to-composition source attribution, inferring content origin and expression origin before composing them into four collaboration types. We propose Dimension-to-Composition Routing (D2C-Routing), which routes content-side and expression-side evidence to supervised dimension heads before a learned gated composition layer predicts the final label. On MixD2C, a reconstructed split derived from the HART mixed-origin benchmark, our disclosed D2C-Routing-based detector system reaches 0.8603 four-way Avg TPR@1%FPR, 6.5 points above the same-split RACE-local rerun. Core ablations support the routing design, while error analysis shows that distinguishing AI-content/human-expression from fully AI-generated text remains the hardest boundary. Code is available at https://github.com/bystander563/d2c-routing-artifact.

This paper introduces Embodied Scene Rearrangement Planning (ESRP), a novel task requiring embodied agents to rearrange furniture in 3D scenes to match a target configuration using only egocentric observations and a top-down target layout. Unlike prior rearrangement tasks, ESRP precludes global state access and introduces mutual object occlusions, reflecting the practical constraints of real-world robotic deployment. These factors make aligning partial egocentric observations with the global target layout particularly challenging for long-horizon planning. To facilitate research, we present ESRP-Bench, a comprehensive benchmark built on OmniGibson featuring over 5,400 scene pairs and 8,200 objects. We define three multi-level metrics to evaluate rearrangement quality and provide four baselines: a hierarchical task-and-motion planning method, a vision-language-model-based method, and two learning-based approaches (IL and RL). Experimental results demonstrate that current methods struggle to complete the task efficiently, highlighting ESRP as a challenging frontier for embodied agents in scene understanding and long-horizon task planning. This work serves as a stepping stone toward deploying intelligent agents in real-world scenarios. Project page: https://pie-lab.cn/ESRP/.

Language model pretraining has become almost synonymous with prohibitive cost, placing it out of reach for much of the academic and open-source communities. Although strong open-source efforts already exist, including open-weight models and open-source training recipes, a cost-efficient, hardware-accessible, and open-source pretraining recipe has long been missing. Even at a small scale, training Llama-3.2-3B costs over \$1.5M, and reproducing SmolLM3-3B needs over \$700K. In this report, we present an open pretraining recipe designed to lower this barrier. Using this recipe, we train a collection of Puro-2B models from scratch on up to 1.4 trillion tokens with FP8 precision on consumer-grade RTX 5090 GPUs. The models in the collection differ in token budgets and selected recipe variants. Our best model is trained at a compute cost of less than \$6.9K and approaches Qwen2.5-1.5B performance under our evaluation protocol. This cost efficiency is enabled by a combination of approaches, including hardware selection, low-precision training, hyperball optimization, curriculum model averaging, and the data recipe. Beyond the recipe itself, we provide two additional results. First, across the Puro-2B collection, we derive a Puro Cost Scaling Law that relates training cost to average model performance; the fitted law suggests that about \$4.4K, less than \$5,090, is sufficient to reach the performance of Qwen2-1.5B. Second, as an end-to-end case study, we examine how pretraining data curricula shape downstream performance after post-training. Such controlled studies are enabled by having access to the full pretraining pipeline rather than model weights alone. We release the full training recipe for Puro-2B, including data, code, and model weights under Apache 2.0 at https://huggingface.co/collections/thu-pacman/puro-2b.

Joint-Embedding Predictive Architectures (JEPAs) for world modeling typically employ fixed-size Vision Transformer encoders that are over-provisioned for simple tasks and under-provisioned for complex ones, with significant redundancy across attention heads. We propose Successive Capacity Growth (SCG), a method that starts from a minimal encoder (1 head, 2 layers, 283K parameters) and grows incrementally in width (adding attention heads for low-level semantic capacity) or depth (adding transformer blocks for higher-order semantic abstraction), driven by a task-agnostic test-and-verify mechanism that exploits function-preserving expansion to safely trial architectural changes and roll back if they do not improve prediction loss. The Sketched Isotropic Gaussian Regularizer (SIGReg) ensures that all learned semantic dimensions remain statistically independent and aligned with the predictive objective, preventing collapse even as the architecture grows. On a 60-dimensional multi-object dynamics task, SCG naturally triggers depth expansion, improving prediction loss by 20.3% over the fixed small baseline with 56 times greater parameter efficiency than scaling to the fixed large model; on a 2D navigation task, a single width expansion yields even an 23% improvement over the fixed large model. Across all three tested environments of increasing complexity, the adaptive encoder matches or exceeds the fixed small baseline, with zero false-positive expansions and bit-exact function preservation (ratio = 1.0, absolute difference = 0.0). The take-away is that JEPA world model encoders need not be pre-allocated at maximum capacity - they can grow successively as the task demands, achieving significant compute and data efficiency while maintaining representation quality.

Speaker anonymization suppresses speaker-identifying attributes from speech while preserving linguistic content and quality. We propose repurposing XTTSv2, a multilingual voice cloning model trained on 27k hours of speech, for speaker anonymization without retraining. Our key insight is that XTTSv2's voice cloning capabilities preserve prosodic structure independently of speaker identity, enabling voice conversion by conditioning on a pseudo-speaker. We introduce an iterative refinement strategy that balances privacy and utility by maximizing a harmonic mean of speaker dissimilarity and intelligibility. Evaluated on seven European languages across CommonVoice and Multilingual LibriSpeech, our system achieves near-optimal privacy (EER $\approx$ 0.49), competitive intelligibility, and substantially better speech quality than dedicated anonymization baselines, while requiring no language-specific training. We release the code here: https://github.com/rm00cr/coqui-tts.

Evolution Strategies (ES) have recently emerged as a memory-efficient post-training paradigm for LLM reasoning. However, the optimization behavior of ES remains understudied, making it hard to define its advantage scope compared to mainstream post-training paradigms (e.g., Group Relative Policy Optimization (GRPO)). By systematically investigating ES dynamics and mechanisms, this paper first identifies a performance advantage of ES over GRPO, theoretically and empirically showing that ES can lead to broader reasoning coverage, thereby better exploiting the reasoning capabilities of pretrained LLMs. Theoretically, we show that verifier-projected Jensen-Shannon diversity across the ES population is helpful to higher Pass@K performances. Empirically, unlike GRPO, which exhibits entropy collapse, ES improves Pass@1 while attaining higher Pass@K than GRPO. We further develop a sequential GRPO-ES training strategy that combines GRPO's strength in Pass@1 with ES's gains in Pass@K. Second, we find that despite substantial whole-model parameter drift, the task-performance gains of ES are only contributed to a sparse subset of larger-magnitude updates. This functional sparsity suggests that large parameter movement need not imply widespread functional change, and held-out evaluations further show that it does not necessarily lead to catastrophic forgetting. Finally, we study how hyperparameter design affects the effectiveness of ES, demonstrating that ES requires a smaller population size in a larger LLM. These findings position ES as a distinct reasoning post-training paradigm rather than a less effective, memory-efficient alternative to GRPO.

As large language models (LLMs) are deployed as autonomous agents, safety failures increasingly involve consequential actions. We study agentic misalignment, where agents take harmful actions under goal conflicts and pressures. Using chain-of-thought (CoT) monitoring, we find that harmful execution is often preceded by intent signals in reasoning. However, post-hoc CoT labels are too coarse to show how intent changes during generation. We introduce INTENT-AS-A-TOOL, an approach that adds intent-targeted tools to give the model a dedicated channel for expressing commitment to a target behavior. The probability of calling an intent tool provides a judge-free, fine-grained signal of the model's tendency to pursue that behavior. Our results show that INTENT-AS-A-TOOL complements CoT monitoring, expands post-hoc CoT labels into dense trajectories, and identifies critical steps for online intervention. These findings suggest that action preferences are useful for tracking agentic misalignment during reasoning. Our code and data are accessible: https://github.com/RebeccaZhang22/intent-as-a-tool.

Recent video generation models are increasingly framed as world models. Many physical processes can unfold in more than one valid way. Therefore, a world model should reproduce not only a plausible trajectory, but also the distribution of possible behaviors under the same initial observation and action. We call this distribution-level requirement probabilistic alignment. However, existing evaluations largely assess individual-video plausibility and do not test whether repeated generations recover the correct distribution. This raises a central question: how far are current video generators from probabilistically aligned world modeling? To answer it, we formalize probabilistic alignment as a distributional criterion for world models and introduce PAWBench, a benchmark for evaluating video generators as stochastic samplers of world dynamics. We further introduce PAWEval, an outcome-level protocol that converts repeated video rollouts into empirical distributions over possible physical behaviors. Across 50 scenarios and eleven current systems, no model consistently matches the reference probabilities while recovering the range of valid behaviors. Having established this gap, we test whether language prompts, initial noise sampling, or model training can reshape the model's predictive distribution. We believe our work can serve as a foundation for future efforts to move towards probabilistically aligned world modeling.

Specializing Large Language Models (LLMs) toward distinct abilities underpins successes ranging from personalized assistants to multi-agent systems (MAS). Single-agent paradigms rely on pre-defined personas or steering vectors to induce specialization, yet they impose a single fixed specialization that fails to adapt to diverse queries. Conversely, MAS achieves dynamic multi-perspective problem solving by orchestrating agents with distinct text-based roles, but fusing these specializations requires multi-turn interactions that inflate context length and inference cost. To address these limitations, we propose Mixture of Roles (MoRe), which adaptively composes multiple specializations into a single steering vector for single-turn inference. Specifically, MoRe learns a diversified codeboox of steering vectors, each of which encodes a latent role. A query-aware router dynamically fuses the codebook into a steering vector that encompasses multiple roles. By steering the backbone LLM with the composed vector, MoRe enables multi-perspective specialization in a single-agent, single-turn inference process. The proposed MoRe can be efficiently trained via a three-stage SFT curriculum and GRPO post-training, while the backbone LLM remains frozen. Experiments across reasoning and personality benchmarks show that MoRe outperforms single-agent baselines by 2.2% on average, and achieves performance on par with MAS while reducing token cost by 20x.

The field of photonic integrated circuits (PIC) has flourished in the past two decades, fueling numerous cutting-edge applications across sensing, networking, data interconnect, and quantum information processing. As a guiding material for PIC, Si$_3$N$_4$ has seen extensive use for its ultra-low loss, broad transparency, and diversity in implementation across both thin and thick films. Although the standard, traditional silicon dioxide (SiO$_2$) on silicon (Si) substrates that underpin the majority of Si$_3$N$_4$ photonics face drawbacks in the form of long-wavelength transparency limited by SiO$_2$, high-stress deposition for anomalous dispersion thick-film Si$_3$N$_4$, and leakage loss to the Si layer for low-confinement thin-film Si$_3$N$_4$. Featuring increased long-wavelength transparency into the mid-infrared, low-stress deposition of Si$_3$N$_4$, and a low index, this work investigates sapphire substrates as alternate hosts for Si$_3$N$_4$ photonics with greater spectral coverage and reduced fabrication complexity. This work presents a robust method of fabricating ultra-low loss photonic integrated circuits on a 500-nm-thick Si$_3$N$_4$-on-sapphire platform, exhibiting record-low losses below $0.1 \rm \;dB/cm$. Implemented using this process are high-Q microrings with intrinsic quality factors in excess of $4.5\times10^6$ and coupled-ring photonic molecules to support nonlinear gain. Leveraging the achievable low loss and high-Q, this work further reports the first demonstration of Kerr-comb and soliton generation on the Si$_3$N$_4$-on-sapphire platform. These advances in loss, quality factor, and soliton generation on this versatile, broad-transparency platform pave the way for future work in spectroscopy and quantum-enhanced sensing across previously prohibited spectral regions for Si$_3$N$_4$ photonics with reduced fabrication complexity.

Industrial sites contain large volumes of read-only telemetry, but few benchmarks specify how to compile these records into executable multi-turn agent tasks. We present a telemetry-to-episode construction method instantiated as BTS-AgentBench. The pipeline normalizes BTS metadata and raw histories into a read-only tool store, compiles static tasks with tool-derived gold answers and evidence, and lifts retained tasks into typed, bounded operator-facing episodes. The 532-row release adds clarification, goal revision, timestamp policy, quality-gated reporting, and evidence attribution while preserving the source computation and split. Coded contract preflight reports zero findings, and the construction-exclusion controller completes 0/532 rows. Two independent raw-to-episode builds match all 11 logical tool-store exports and reproduce the released 356/87/89 train/dev/test artifact exactly. Applying the shared construction path to XAI4HEAT produces 204 episodes; on its 41-row held-out test split, the controller completes 0 rows and the retained GPT-5.5 execution completes all 41. Code, artifacts, and replay reports are available at https://github.com/kjy7567/BTS-AgentBench.

High similarity between first-visit and return frames does not necessarily show that a video world model remembered the scene; the intervening rollout may simply have changed very little. This ambiguity makes absolute revisit scores sensitive to rendering stability, repetitive content, and failed motion. We introduce \emph{R2M-Bench} (\textbf{R}elative \textbf{R}evisit \textbf{M}emory Benchmark), a benchmark of observable revisit-selective consistency. For every detected return, R2M-Bench compares the revisit pair with two controls from the same rollout: a gap-matched non-revisit pair that measures generic temporal stability and a short-range pair that estimates short-horizon consistency. These comparisons produce \emph{MemoryGain} (MG), the revisit advantage over the temporal baseline, and the \emph{Normalized Memory Ratio} (NMR), which normalizes this advantage by the short-to-baseline dynamic range. R2M-Bench combines 100 reference scenes with three leave-and-return trajectories to form 300 instances and evaluates appearance fidelity, scene and object identity, local geometry, and persistent state. Across seven action-conditioned video world models, Overall NMR correlates with human consistency judgments at Spearman's $ρ=0.547$ (95\% CI $[0.45,0.63]$). Its within-model correlation magnitude with generated motion is $0.072$, compared with $0.207$ for raw revisit similarity, indicating that relative calibration substantially reduces the slow-motion shortcut. DreamX-World-Memo achieves the highest Overall NMR among the evaluated video models. Together, these results support same-rollout relative calibration as a practical way to distinguish revisit-specific consistency from generic temporal stability.

Agent harnesses shape how language-model agents use instructions, tools, and runtime components, but adapting these harnesses requires costly verification. Existing propose-and-verify methods typically score every candidate on a fixed task set, wasting rollouts on unrelated behaviors and allowing aggregate scores to obscure specific regressions. We introduce HarnessLens, a budget-aware framework for automated harness evolution. HarnessLens jointly explores the task space and user-configurable components, derives candidate modifications from execution trajectories, and selectively verifies each candidate on behavior-relevant tasks using an attributable-evidence gate. Across three agent harnesses and four benchmarks, HarnessLens improves average held-out performance by 7.6-13.6% while consuming substantially less evaluation budget than competing baselines. These results demonstrate that behavior-aware verification with explicit attribution enables more reliable and sample-efficient harness evolution under constrained interaction budgets. Our code is available at https://github.com/jhxu5214/HarnessLens.

Accurate identification of the correct view or angle in cardiac ultrasound (echocardiogram) is a critical component of cardiologic imaging. This step is essential for precise anatomical interpretation, reliable measurement, and the reduction of clinical errors. Although computer vision has advanced significantly, most state-of-the-art models perform well on standard benchmarks but often yield suboptimal results in specialized medical imaging tasks due to the high level of noise present in the data. QuantumBoostNet, a hybrid classical-quantum architecture, is introduced to address these challenges. This model integrates a classical backbone with two heads: one classical and one quantum, with the quantum head implemented as a parametrized 10-qubit quantum circuit. Training occurs in two stages, with an adaptive transition between heads governed by a mixing parameter that monitors loss dynamics. Extensive experiments indicate that, despite the limited number of qubits that can be simulated, QuantumBoostNet consistently outperforms state-of-the-art classical and hybrid classical-quantum models in cardiac ultrasound view identification, achieving a relative improvement over the best competitor. QuantumBoostNet also demonstrates superior performance on established image classification benchmarks and exhibits robustness to noise. These findings support the continued development of hybrid classical-quantum models for specialized medical imaging applications.

Most single-stage 3D object detectors complete different tasks with the same extracted features. Nevertheless, it is impossible to project features into a common space that is adaptive for all the tasks. We present a novel task-aware deformable prediction (TADP) method for single-stage 3D object detection to solve this problem. Firstly, a triple feature refinement aggregation module is designed to extract three-level features adaptively. Additionally, we design the multi-scale feature aggregation block to fuse multi-scale features in a scale-aware manner. Finally, the prediction of each task is deformed with the designed plug-and-play task-aware deformation head. It can percept the emphasis and interaction of each task. We also designed three different deformation modules. The experimental results demonstrate that the proposed deformation head shows good results on other detection methods. The experimental results on the KITTI dataset demonstrate that the car mAP is 80.91%, surpassing many state-of-the-art methods on the KITTI benchmark.

Although Large language models (LLMs) mediate access to knowledge and computational assistance, their capabilities should benefit vulnerable groups in the same way. However, it is unclear whether existing AI systems are inclusive enough for blind and deafblind users to access the same functionality through Braille, whose indicators, contractions, and digital representations introduce distinct requirements for model comprehension. To this end, we introduce BrailleBench, a benchmark for evaluating LLMs in Braille comprehension from different Criteria. BrailleBench aligns 5,570 instances from five datasets, including mathematics, commonsense, and multi-hop question answering across English and Braille Grades 1 and 2. Different configurations are designed to understand whether the systems can comprehend Braille-authored content, express answers in Braille, and complete end-to-end Braille interaction. To ensure the quality and prevent evaluation bias, the benchmark is built through a deterministic, expert-reviewed pipeline via a self-created Braille Toolkit without using any data instances generated by LLMs. We evaluate six representative LLMs from various aspects. The results reveal a persistent gap between print-English capability and Braille accessibility. Braille understanding and expression are asymmetric, where Grade 2 is especially fragile on the input side compared to Grade 1, and fully Braille requests further reduce performance. The experimental observations provide valuable guidance for the development of future Braille AI systems. All related resources in BrailleBench are publicly available for future research.

Efficiently improving autonomous agents across diverse tasks is central to accelerating recursive self-improvement (RSI) in agentic AI, with prompt optimization emerging as a promising approach capable of delivering performance gains comparable to those achieved by fine-tuning model weights, while reducing computational costs in both optimization and serving. However, recent developments increasingly favor unnecessarily complex prompt optimizers. We introduce Naive Prompt Optimization (NPO), a lightweight single-lineage method that iteratively revises prompts using a teacher model with rollout feedback. NPO achieves comparable or better performance than GEPA with fewer rollouts, and its advantage increases with stronger teacher models, suggesting that stronger teacher reasoning can partially substitute for optimizer-side search complexity. In interactive games, NPO remains broadly competitive with GEPA, while GRPO performs better on some tasks less amenable to prompt optimization. We also show that NPO-optimized prompts elicit similar performance improvements when applied verbatim to other student models, especially across models within the same family. Overall, our preliminary results show that simple, linear prompt optimization can rival substantially more sophisticated and complex search procedures.

LLM agents increasingly rely on generated interaction data to learn how to interact with external environments. Agentic data generation must maintain consistency among environments, tasks, interactions, and success signals while producing experience that is useful rather than merely abundant. Existing work spans many agent domains, but domain-centered organization and heterogeneous evaluation often obscure common generation mechanisms and conflate candidate construction with verification and selection. This work develops a two-level framework for the field. First, we represent agentic data as a common factorized object $(E,q,τ,v)$, comprising an environment specification, task signal, interaction realization, and optional verifier. We organize generation paradigms by their primary anchor and dependency structure. Second, we formulate generation as constrained distribution design through the Accuracy-Complexity-divErsity (ACE) lens. Accuracy establishes the feasible support of grounded and internally consistent data. Within this support, Complexity places learning mass relative to the capability of a declared learner and execution configuration, while divErsity controls coverage and redundancy of data. Using this framework, we explore how prior work verifies generated experience, constructs and calibrates difficulty, and expands behavioral coverage. The literature reveals a shift toward execution-grounded accuracy, learner-relative complexity, and diversity beyond surface variation or dataset size. We further discuss broader directions and emerging trends in agentic data generation through the ACE lens, including their implications for scaling, data sources, training regimes and adaptive learning. Overall, the central challenge is not simply to generate more data, but to continually allocate valid, informative, and non-redundant experience as agents and environments evolve.

One approach to mechanistic interpretability explains behavior through circuits: the components and connections that carry it. Frozen discovery often returns hundreds of edges, making them hard to inspect, compare, or verify exhaustively. We introduce Circuit Condensation, which post-trains models to concentrate behaviors into smaller causal graphs. Each round prunes low-attribution edges and trains a low-rank adapter to match the original through what remains, retaining the cut only if task performance and general capability survive. Across four behaviors and eight models, condensed circuits are smaller than the strongest frozen baseline in 30 of 32 settings, by $8.1\times$ on average and up to $316\times$. Repeating the search without weight updates produces larger circuits in 29 of 32 settings, showing that weight updates, rather than search alone, drive the reduction. Testing every subset of 19 circuits finds 11 that cannot be reduced and reveals removable edges in the rest. Pair ablations expose dependencies between edges, showing that their effects cannot be understood independently. On indirect object identification, condensation isolates 24 heads, 17 of them with documented roles, against 61 heads and 36 undocumented ones for the matched frozen circuit: a sufficient sub-circuit of the published mechanism rather than a reconstruction of it. The resulting circuit tracks the original model's next-token distribution and predicts its errors.

An LLM call in a semantic data processing system is expensive enough to dominate query cost, yet slow enough to hide a CPU-side learner's update behind its round-trip. In production, LLM compute accounts for $80-90\%$ of query cost, and each call costs $10^5-10^7\times$ a relational predicate. The latency window inverts a design constraint of classical adaptive query processing, where online learners had to stay lightweight to avoid dominating the predicates they optimize. At LLM latency, per-call gradient steps and per-batch threshold solves fit inside the round-trip. We develop compositional online learning at the LLM call boundary: a framework for combining online-learning components in semantic data processing systems. Each component makes execution-time decisions and refines its learned artifacts online. The design space spans two axes, decision granularity and learner update cadence, and the components share a single learning pattern that hides each trainer step inside the next LLM round-trip. A production case study in Cortex AISQL composes three components: a memoization layer, an online per-call filter-ordering learner, and an online per-batch cascade-routing learner. A conditional cost decomposition assigns each learning component to a distinct factor of per-row LLM cost. Under independence, the two learning components compose multiplicatively to an $11.4\times$ upper bound on a representative conjunction-filter workload. Self-selection at the cascade boundary, sample-budget shrinkage, and selectivity-estimation drift reduce it to a realistic figure near $8\times$.

We introduce a self-supervised framework for learning joint visuospatial representations from RGB-D observations. While modern vision foundation models are trained almost exclusively on RGB images, many embodied systems have access to explicit depth sensing, which provides geometric information that monocular inputs cannot recover. Our method integrates depth-derived geometric priors with a visual backbone through inter-patch and intra-patch fusion, enabling the model to encode both appearance and spatial structure efficiently. The resulting representation shows promising improvements on 3D awareness while preserving semantic transfer: it outperforms prior methods of comparable scale on multiple 3D geometry benchmarks, and remains competitive when probed for standard RGB-D semantic segmentation tasks.

Robots that can change their morphologies and behaviors for different tasks and environments hold great promise for adaptable, multifunctional systems. Modular reconfigurable robots (MRRs) can achieve such functionalities by docking and rearranging individual units, but most rely on rigid modules that lack structural compliance, resulting in limited capabilities. Continuum robots offer compliance through flexible backbones, yet they cannot self-reconfigure into task-adaptive multi-robot configurations. Here, we introduce an MRR that unifies the advantages of both architectures by combining a tensegrity-based compliant body with claw-based connection mechanisms. Each robot can manipulate and locomote independently, and multiple robots can self-reconfigure into different morphologies (e.g., chains, loops, branches) for cooperative manipulation and locomotion. We demonstrate the robots' capability across diverse tasks and environments, including coordinated object manipulation and transport, multimodal locomotion, and loco-manipulation in real-world scenarios. These results lay a foundation for adaptable and multifunctional robotic collectives, with broad potential applications in manufacturing, space exploration, and search-and-rescue operations.

Mental health assessment relies on episodic self-report scales, which convert subjective states such as stress into numerical scores but provide only sparse snapshots of wellbeing. Wearable devices offer longitudinal behavioral and physiological signals for continuous, low-burden monitoring. Recent LLM-driven personal-health agents enable natural language queries over wearable signals, but mainly handle short-term, retrieval-based lookups (e.g., highest step count over a week). They do not evaluate whether agents can reason over long-term signals to predict wellbeing scores paired with evidence-grounded rationales. To address this gap, we introduce BALMS, the first systematic benchmark of LLM-based agentic systems for longitudinal mental health sensing. BALMS spans 3 real-world longitudinal datasets, 2 task families (closed-form wellbeing-score prediction and rationale generation auto-graded by an LLM-as-Judge), 3 agentic paradigms evaluated across 5 open- and closed-source LLM backbones. We find that zero-shot agents rarely outperform a simple mean baseline, except with stronger backbones or compact, semantically meaningful features. Chain-of-thought prompting improves reasoning-oriented backbones, but does not guarantee temporal grounding or numerical correctness. Together with more analysis on efficiency and temporal scaling, BALMS highlights the need for longitudinal mental health agents that selectively retrieve history, ground temporal evidence, and reason over interpretable behavioral features.

Open World Object Detection (OWOD) built on multimodal foundation models often suffers from semantic ambiguity caused by unidirectional text-to-vision matching, while rigid outlier penalties may over-suppress unknown objects near known-class decision boundaries. We propose CODE (Cross-Modal Calibration and Dynamic Suppression), a unified inference-time framework with three complementary components. Cross-Modal Joint Confidence Calibration injects global visual prototypes to calibrate text-driven known-class predictions. Uncertainty-Guided Universal Objectness Enhancement measures classification hesitation from local visual responses to strengthen potential unknown objects. Dynamic Outlier Suppression via Confidence Margin replaces rigid suppression with a margin-aware adjustment that preserves ambiguous out-of-distribution instances. Experiments on the Real-World Detection benchmark demonstrate that, with the OWL-ViT L/14 backbone, CODE achieves 21.7 U-mAP and 40.8 K-mAP in Task 1, surpassing the previous state of the art by 2.6 and 2.3 points, respectively.

Nitrogen rates for winter wheat are set before the season, under unknown prices and weather. The standard UK advice does not respond to prices, yet recent price swings moved the most profitable rate by tens of kilograms per hectare. Machine learning is often proposed as the fix. However, it is usually judged on prediction accuracy, and accurate prediction does not by itself make the recommended rate more profitable. Our insight is to score nitrogen advice directly by the profit it forgoes on measured yield response curves. We build a test bench on 892 such curves from two long running UK experiments, and sweep the nitrogen to grain price ratio to cover all price scenarios. On this bench, machine learning fails as a predictor. No model recovers the best rate within farm tolerance, and the benchmark noise shows none can. At normal prices, every model also loses to the standard advice on profit. The gain sits elsewhere. A simple correction step applied after the model cuts profit losses by a quarter, while better models and extra features give no gain. The same frozen correction cuts losses by 43% at the second site without any retraining. A hybrid of standard advice plus a damped correction removes bias and trims rare large losses. The same price sweep also prices emission cuts, at a cost comparable to current carbon prices. Machine learning therefore pays as a profit scored correction to standard advice, not as its replacement.

Accurate 3D segmentation of cone-beam CT (CBCT) is critical for interventional and radiation therapy applications, yet it remains limited by two compounding challenges: the scarcity of annotated CBCT data and the large domain shift from diagnostic CT. Interventional CBCT exhibits fundamental modality differences from conventional CT, driven by acquisition and physics effects as well as contrast-specific vascular content, thereby limiting effective cross-modality model transfer. We propose a novel unsupervised domain adaptation (UDA) framework based on redundancy-reducing feature alignment, enabling 3D CBCT segmentation with no target-domain annotations or inference-time adaptation. Our framework is architecture-agnostic, seamlessly adapting both CNN-based and ViT-based foundation models. We evaluate our method on two challenging CT-CBCT liver segmentation benchmarks: one for interventional vascular procedures and one for radiation therapy, demonstrating that even large-scale pretrained segmentation networks require explicit feature-space bridging to generalize across acquisition modalities, and that our approach consistently outperforms existing pretrained foundation model and UDA strategies. To support reproducibility and benchmarking, we release the liver segmentations for a public CBCT dataset, along with the code, trained models, and weights.

Peer effects are difficult to estimate when interaction graphs evolve because pre-assignment network history, dynamic peer exposure, and post-assignment network change have distinct causal roles. We introduce a controlled contrast framework that indexes potential outcomes by own treatment, temporally aggregated peer exposure, and a post-assignment evolution summary. Differences between the resulting means define own-treatment, peer-exposure, controlled network-evolution, and joint controlled contrasts rather than a mediation decomposition. We develop the Dynamic Network Doubly Robust estimator, DynaNet-DR, which combines a temporally factorized propensity with normalized augmentation. Under consistency, summary sufficiency, sequential exchangeability, positivity, nuisance convergence, and weak dependence, its canonical estimator is consistent when either the outcome regression or the propensity estimator is consistent. The reported implementation adds representative-score prediction, fixed clipping, and finite-sample stabilization. Semi-synthetic benchmarks on fixed real temporal graph sequences show favorable estimation accuracy among methods targeting the full profile. These benchmarks assess summary-indexed contrasts rather than counterfactual edge generation, and the MathOverflow study is an observational illustration under the stated assumptions.

Soft actuators enable dexterous and compliant interaction, but closed-loop task-space control remains challenging due to strong nonlinearities, distributed deformation, and uncertainty in their dynamics. This paper presents a real-time dynamic-model-based task-space feedback and estimation framework based on a non-minimal coordinate discrete elastic rod model formulated in absolute coordinates with holonomic constraints. The resulting structure preserves distributed mechanics while maintaining computational efficiency through sparse system matrices, enabling real-time control with up to 10 discretized rods. A quasi-static feedforward inverse model is combined with a task-space PI controller and a dynamic observer that fuses measurement residuals as virtual forces, enabling full-state estimation from sparse sensing. The approach is experimentally validated on three planar pneumatic soft actuators with varying geometries. Across five tasks, including drawing the digits 0-9 across the workspace (3-18 mm/s tip speed), tracking periodic motion (up to 37 cm/s), cross-platform generalization, reduced sensing conditions, and real-time user-defined references, our method achieves 1.5-2.3 mm root mean square error (RMSE) for precision motions and 5.5-12.4 mm RMSE at 1-2 Hz. Results demonstrate that structured, non-minimal dynamic models can enable real-time, high-precision, moderate-bandwidth task-space control of planar soft pneumatic actuators in free space.

Electronic design automation (EDA) has advanced engineering productivity through successive generations of tooling that progressively automate synthesis, optimisation, and verification. Large language models (LLMs) extend this trajectory by enabling direct translation from design intent to hardware implementations. In most of the EDA literature, LLM-based solutions are typically assisting siloed design stages or tasks, however this obscured the drivers by which capability emerges and systems scale. In this Perspective, we instead define three hierarchical roles that reveal how capability accumulates: a Generator that produces design artifacts in a single pass, an Agent that refines outputs through iterative tool feedback, and an Orchestrator that coordinates decisions across EDA-stages. Across published systems, this reveals a syntax trap in which models are trained to produce plausible code rather than physically correct hardware, compounded by fragmented tools and loss of design context that obscure how decisions affect later stages. Comparisons across the three roles show that current approaches struggle to scale to industrial designs, motivating a shift towards a standardised, physics-aware orchestrator that connects tools and agents across the EDA flow for more reliable and accessible hardware design.

LLM agents are increasingly applied to anomaly detection and root-cause analysis in time-series observations collected from real-world systems; however, their performance on these tasks has not been systematically evaluated under controlled conditions. We introduce TraceBench, a simulation-based framework for generating controlled root-cause attribution tasks. In each generated task, an agent receives time-series observations produced by simulating a physical dynamical system and must determine whether a system parameter was altered during the simulation and, if so, which one. Using TraceBench, we generate tasks from three interpretable mechanical systems and systematically evaluate four LLM agents across controlled experimental conditions, yielding new insights into how these agents analyze time-series observations from dynamical systems. Our results show that agents benefit substantially from domain context and explore data primarily through numerical console output rather than visualizations. We also find that agents generally perform worse when required to produce a Python script that maps each time-series sample to a predicted root-cause label than when they submit predictions directly. We release our datasets, agent trajectories, experimental results, and a leaderboard on our website, tracebench.github.io.

Understanding spoken dialogue requires joint reasoning over lexical content and paralinguistic acoustic signals such as emotion and conversational intent. However, existing evaluations often allow shortcuts based on transcripts or single-modality solutions, obscuring whether models genuinely ground predictions in speech. We formalize this failure mode as cross-modal disagreement, where transcripts suggest plausible but incorrect surface interpretations while acoustic cues such as prosody or speaking style support different answers. We develop a scalable framework that identifies text-biased surface interpretations and converts disagreement regions into conflict QA examples. We also include consistent cases where transcript-based and speech-grounded interpretations agree, enabling evaluation beyond adversarial audio dependence. This results in ContraTalk, a controlled benchmark containing 501 questions across five discourse dimensions: interaction behavior, emotion state, dialogue act, social stance, and conversational intent. We further develop an agentic-style reasoning framework that converts speech into an Audio Twin, a text-readable representation of localized acoustic cues that exposes acoustic evidence to the reasoning model. Experiments show that strong text-only LLMs exceed 90% accuracy in consistent cases but drop to 33-48% in conflict cases. Direct AudioLLMs provide only partial grounding, still selecting the transcript-biased trap in roughly 30-40% of conflict cases. Our Audio Twin framework improves conflict-case accuracy while reducing trap selection, but its consistent-case behavior remains backbone-dependent. These results identify transcript-based shortcuts as an important failure mode in spoken dialogue understanding and show that explicit acoustic evidence aggregation provides a more controllable interface for diagnosing and improving speech-grounded reasoning.

Many Earth Observation applications need land-use/land-cover maps that are both precise and frequently updated, yet the strongest Earth Observation foundation models build their embeddings from a full year of observations. We present a controlled study of the temporal sensitivity of Tessera, one of these leading foundation models, for land-use/land-cover mapping. Keeping the encoder frozen, we recompute its embeddings over varying observation windows, from a full year down to a single day. We use them as inputs to a linear probe and a UNet segmentation head, benchmarking both of them against from-scratch networks on LUCAS, DynamicEarthNet, and PASTIS-R datasets. We show that the value of the embeddings is task-dependent. Where classes are separated by phenology, as for the crop types of PASTIS-R, they reach a mean Intersection-over-Union of $58.3$, about $46\%$ above the best from-scratch model. Where classes are temporally stable (e.g., forests in DynamicEarthNet and LUCAS), embedding-based and from-scratch models match only under full supervision. On both datasets, Tessera embeddings remain markedly more label-efficient. Degradation under shorter temporal windows is gradual and class-dependent. Contracting the window from one year to one month costs $39\%$ of the segmentation accuracy on PASTIS-R but only $5\%$ on DynamicEarthNet. Single-day embeddings still classify land cover in LUCAS at $3.4$ times the chance level. Our study shows that temporal coverage is therefore a tunable cost rather than a fixed prerequisite, opening regimes such as near-real-time mapping and faster land-use/land-cover refresh cycles.

Ancient Chinese artifact text recognition is fundamental to heritage digitization, and benchmarks for ancient texts are essential for evaluating current model capabilities. However, existing benchmarks suffer from ''fragmentation'', manifested in limited temporal coverage, limited medium diversity, and incomplete script types. Therefore, we present Ancient-Bench, a comprehensive benchmark of 2,700 images for ancient Chinese artifact text recognition, featuring three dimensions: Multi-millennial (spanning 3,000 years of character evolution), Multi-medium (covering nine artifact categories), and Multi-script (encompassing seven historical script forms). To enable consistent and fair evaluation across heterogeneous media, we further define three annotation standards tailored to the medium-specific characteristics of ancient texts: symbol standardization, character standardization, and parsing standardization. Extensive experiments on Ancient-Bench covering general Vision-Language Models (VLMs) and OCR-specialist models reveal that ancient Chinese artifact text recognition remains fundamentally unsolved, with persistent challenges in variant characters, specialized symbols, and hallucination. The dataset is available at https://github.com/SCUT-DLVCLab/Ancient_Bench.

An LLM agent shown a professional-looking market panel commits to a directional call on a provably unpredictable question far more often than one asked the bare question: across 12 frontier models, commitment rises from 6.5% to 54.0% as evidence is escalated. It commits just as readily when every number on the panel is invented: fabricating the entire display, so nothing the model can see is true except the question itself, still lifts commitment from 24.5% to 36.8%, statistically indistinguishable from the 37.6% produced by genuine market data. What unlocks confident action is not information but the authority of its packaging. The failure is narrow and locatable. Incapacity is not the answer: on matched answerable questions attached to the same panels, the same models answer essentially always, at near-perfect accuracy. Nor is it belief - stated probabilities barely move across the gradient that swings action by 48 points, and score worse than a climatological baseline. Missing judgment isn't it either: asked to classify a question's knowability before acting, models call it irreducible 90% of the time and then commit on just 0.4% of those. The act/don't-act gate is what fails, and the effect is concentrated in a few models rather than universal. Because the gate is separable, it can be trained. Supervised fine-tuning of a 3B model on 540 synthetic cases, predominantly dice, coins, jars and timers, drives commitment to 0.0% on the original cases and transfers to three unseen domains. It does not survive everything: the gate holds exactly when the response format leaves room to reason, and rigid formats that remove that room leave the model confident and wrong on questions it otherwise answers correctly. The gate is trainable and context-fragile, and deployment needs both halves of that sentence.

Autoregressive large language models (LLMs) routinely generate factually incorrect outputs with high decoding confidence, limiting their deployment in high-stakes workflows. Existing output-stage uncertainty metrics can fail when models are overconfident on false assertions, while multi-sample verification pipelines introduce substantial memory and latency overhead. This work evaluates whether internal hidden-state transition dynamics during generation can signal factual errors without auxiliary decoding calls. We introduce Prediction of Prediction (PoP), a mechanism that captures layer-transition uncertainty by fusing intermediate hidden representations across depth during a single forward pass. Evaluated on the TruthfulQA benchmark using autoregressive transformer backbones, PoP achieves an area under the receiver operating characteristic curve (AUROC) of 75.5% for factual-correctness classification. The mechanism operates within the base forward pass, adding less than 1.2% runtime latency and requiring zero additional generation passes. The numerical results are reported from the author-verified experimental implementation and are bounded by the evaluation scope described below.

Robot crowd navigation requires safe and efficient decision-making under dense, dynamic, and multimodal human--robot interactions. Existing reinforcement-learning methods typically output a single reactive action at each timestep, which limits their ability to represent diverse short-term avoidance strategies. We propose Planning Diffusion Policy Optimization (PDPO), an offline-to-online reinforcement-learning framework that uses a diffusion policy to generate short-horizon action chunks for crowd navigation. PDPO is first pretrained on collision-avoidance demonstrations and then fine-tuned online with PPO by treating the denoising process as an internal decision process. During execution, the policy generates a five-step action chunk and applies it in a receding-horizon manner. Furthermore, we observe an evaluation artifact in common crowd-navigation benchmarks: without explicit boundary constraints, learned agents may leave the valid domain and bypass dense crowds. To address this, we introduce a setting in which boundary violations are treated as collisions. Experiments show that PDPO obtains an improved success rate over strong baselines, and ablations demonstrate that action chunks are especially important for the modified bounded benchmark.

Vision Language Models (VLMs) have shown great success in general visual tasks, yet they still struggle to deeply understand text within images. In this paper, we introduce ReViCo (Real Visual Correction), a benchmark designed to evaluate VLM text understanding through a novel task of visual text error correction. ReViCo challenges models to identify and fix text errors in real-world images, which requires a profound understanding of the interplay between visual text and its surrounding visual context. We benchmark various VLMs using two distinct paradigms: prompt-based strategy and targeted model training, both aimed at pushing the limits of current models. Our experiments reveal a striking performance gap between even the best VLMs and human, and further analysis also shows that most models struggle to accurately perceive the visual text, resulting in frequent correction errors. By highlighting these gaps, ReViCo provides a new benchmark foundation for developing more robust and text-aware VLMs.

Correlation Clustering (CC) is a natural formulation of clustering in combinatorial optimization, which uses a graph representation of the input and does not require a pre-specified number of clusters. Given $n$ objects and a pairwise similarity function, the goal is to cluster the objects so that similar objects are put in the same cluster and dissimilar objects are put in different clusters. Despite its versatility, existing CC algorithms suffer from significant scalability issues and are inherently transductive: i.e., the algorithm must be executed from scratch for any new problem instance. In this work, we bridge this gap by leveraging Graph Neural Networks (GNNs) to solve Inductive Correlation Clustering, a novel generalization of the CC problem designed to handle unseen graph instances. By learning to exploit common structural patterns and node features during training, our framework generalizes to new graphs drawn from the same distribution with minimal computational overhead with respect to standard algorithms. We demonstrate the effectiveness and scalability of our approach through extensive experiments. Our framework not only excels in the inductive setting, e.g., lowering the inference time up to $5$ order of magnitude, while maintaining an approximation ratio within $~10\%$ of the best baseline solution, but also achieves competitive results on standard (transductive) CC benchmarks. Finally, we showcase a practical application of our framework as a learnable pooling mechanism for graph classification. Our results indicate that our method serves as an efficient pooling layer, enhancing the ability of GNNs to capture hierarchical structural information in networks.

Object classification in event-based computer vision is a task that is attracting considerable research attention. Event-based object classification is a fundamental task in the fields of security and applied computer vision, which typically use synchronous frame-based cameras and computing pipelines for operation. This approach has several practical flaws. The size, weight and power consumption of the device could prohibit deployment at the extreme edge or in covert sensing environments. Besides this, there are security concerns inherent in cloud-based or other off-device computation approaches due to the requirement of sending and receiving potentially sensitive data. Furthermore, this transmission of data introduces latency and requires consistent connectivity to the cloud infrastructure to function. The use of Spiking Neural Networks (SNNs) hosted on neuromorphic devices attempts to solve several issues present in this conventional approach. Research into event-based object classification methods are hindered by the lack of high-quality vision datasets to use. To this end, the ANTShapes simulation tool has been previously proposed to create and label event-based vision datasets. In this paper, four novel datasets of varying difficulties are created using the tool and are benchmarked against existing spiking datasets commonly used for event-based vision research (N-MNIST, CIFAR10-DVS, DVSGesture and POKER-DVS). Classification is performed using a convolutional SNN. This work simultaneously provides four datasets with rich details for future experiments to use and validates the output of the ANTShapes dataset simulation tool as being suitable for its purpose.

The development of frontier models is commonly perceived to be the exclusive remit of a small number of heavily funded players, creating an information, economic and power asymmetry between developers and the diverse user base of modern AI. Recent public discourse acknowledges this concern, calling for SovereignAI (an organisation's capability to independently build, deploy and govern AI use), but offers little concrete advice on how this can be achieved in the short term under a diversity of funding settings. We argue that frontier performance is achievable by a wide range of institutions through Continual Learning on readily available open-weight models. Unlike limited approaches such as small-scale fine-tuning, prompt engineering, or tool-augmentation of a frozen model, our approach exploits a modern mid- & post-training stack while introducing safeguards that preserve both plasticity and stability at each stage, making the minimal number of high-impact interventions on the parameters. This yields gains comparable to those typically seen across multiple successive model generations, at compute and personnel budgets substantially lower than commonly thought, making ownership of large parts of the SovereignAI stack (model, tool infrastructure, values & data privacy) viable for far more actors. We demonstrate this with Thomson, a general-purpose frontier model trained with an enhanced focus on high-stakes professional work. Thomson performs competitively with recent frontier models across agentic tasks, safety, legal, tax & multilingualism, and large-scale Deep Research. Evaluations show a distinctive $π$-shaped pattern: distinct improvements across a wide range of capabilities, including those not explicitly targeted, while almost completely eliminating the forgetting problem common to narrow domain adaptation.

Tool-augmented LLM agents must rely on untrusted runtime Observations to complete open-ended tasks; however, when tool outputs no longer merely provide data but begin to specify concrete actions, they effectively become ``commands'' that can drive real-world side effects beyond user intent. We argue that this risk arises from conflating action induction with execution authorization. To address this distinction, we propose SARA, which treats action induction and execution authorization as distinct runtime roles and separates action provenance from execution authority. On the Observation side, a context-isolated Action Probe exposes action-inducing semantics and persistently records action-origin provenance across steps as a review signal; on the execution side, actual tool calls are authorized only against the user objective and audited evidence from authorized successful executions, while satisfying goal, execution-chain, and argument-level support. To preserve this separation across multi-step execution, SARA applies No-History-Promotion to prevent historical recurrence from laundering action origins into execution authority. Across AgentDojo and AgentDyn, SARA limits ASR to no more than \(0.63\%\) across four primary evaluation settings while maintaining competitive task utility, and consistently reduces ASR across additional Agent backbones.

Despite their potential in standardized graph tasks, Large Language Models (LLMs) remain brittle to real-world shifts in node identifiers and task formulation. While deterministic graph tools are invariant to such shifts, extracting topological structures from noisy text is highly fragile for LLMs, which often overfit to surface patterns. Moreover, mitigating these parsing failures via multi-agent systems incurs prohibitive latency. To address this, we propose GRAIN, a single-agent framework optimized via reinforcement learning. GRAIN models reasoning as a semantic parsing and tool-execution pipeline, guided by a Structure Invariance Reward. By validating extracted intermediate graphs against ground-truth topologies, this reward forces the LLM to learn robust text-to-structure mappings rather than memorizing linguistic artifacts. We also introduce GRIT, a benchmark evaluating sensitivity to such linguistic shifts. GRAIN outperforms multi-agent baselines by 16.45\% in accuracy with approximately 24\% lower latency. Furthermore, it demonstrates superior structural generalization, halving the out-of-distribution (OOD) gap of SFT models (from 15.77\% to 7.80\%) and maintaining robustness on large-scale graphs beyond the training distribution.

Large language model agents are increasingly deployed as autonomous loops. Starting from one human goal, such a system repeatedly discovers work, plans, executes tool calls, verifies outcomes and persists state across many unattended iterations. The agent safeguards in wide use, however, are defined over a single trajectory, and their safety state is re-initialized when the next trajectory begins. We show that this is a failure of composition rather than an implementation detail. Our central result is a separation: against an attack whose evidence is fragmented across several iterations, every trajectory-scoped monitor has a true-positive rate equal to its false-positive rate, however expressive it is, because the evidence it would need never appears in the window it sees, whereas a monitor retaining cross-iteration state separates the two perfectly. We further show that the obvious repair of carrying a geometrically decaying risk score is insufficient, because the cooling-off period a patient adversary must wait is a constant that does not grow with the horizon $N$. We then present LoopHarness, which restores a persistent, non-decaying safety state at the loop level. Under mediated commits and an arbiter detection floor $δ_M$, it bounds the expected number of unauthorized irreversible actions by $B+m-1+m/δ_M$, a constant in $N$, of which the $B+m-1$ term is decided by a model-free rule and therefore survives a fully colluding verifier. We give a complete evaluation protocol on native Agent-SafetyBench tasks with paired clean and attacked episodes, an outer-state attack suite whose decisive evidence exists only across iterations, per-module ablations, and an adaptive white-box red team.

Multimodal foundation models are increasingly used in speech-first assistants that must interpret spoken queries and produce visually grounded decisions. Yet it remains unclear whether semantically equivalent queries yield consistent judgments across modality (text vs. speech) and language (English vs. Arabic). We introduce a speech-augmented visually grounded contrastive triplet benchmark spanning 10,150 culturally grounded images from 18 MENA countries, where each image is paired with one supported statement and two plausible but unsupported alternatives. We define contrastive instability as the conditional rate at which a model fails to resolve all statements within a triplet, isolating fragmented reasoning from complete failure. Evaluating recent multimodal models under text and speech in English and Arabic, we find that modality and language shifts introduce substantial triplet-level inconsistencies that are not fully captured by aggregate accuracy, with speech amplifying partial failures. We make the benchmark publicly available to the community.

Long-context inference is bottlenecked by the memory footprint of the key-value (KV) cache, especially for small models under tight resource budgets. Existing KV cache eviction methods score tokens using the model's attention distribution or, in attention-free variants, each key's distance from a global reference point. Using a controlled leave-one-out probe, we find that attention magnitude is unrelated to a token's causal contribution to the answer (Spearman $ρ=-0.004$), challenging the premise behind dominant eviction methods. We introduce TwinKV, a training-free, attention-free redundancy signal that detects whether a token's key has a near-duplicate elsewhere in context. Rather than replacing existing policies, TwinKV acts as a composable repair pass: given a policy's fixed retained set, it identifies evicted tokens with no surviving duplicate (\emph{orphans}) and retained tokens whose information is duplicated elsewhere (\emph{redundant donors}), then swaps them while preserving the original budget and scoring rule. We compose TwinKV with four recent eviction policies across LongBench, LooGLE, RULER, and a short-context MMLU-Pro no-harm control at compression ratios ${0.3,0.5,0.7}$. On Qwen3-4B, TwinKV improves a majority of configurations for two policies, is near-even for a third, and helps only a minority for a fourth adaptive baseline already near a performance ceiling; gains across the three non-ceiling policies are smallest at the loosest ratio. On RULER with Llama-3.2-1B, however, that fourth policy improves in every evaluated cell because its Alone score leaves substantial room to improve. More broadly, Llama-3.2-1B shows a smaller average LongBench gain but a higher fraction of improved cells on LongBench and LooGLE than Qwen3-4B, plus a clean RULER win. We also identify few-shot classification exemplars as a task structure where TwinKV does not help on either model.

Internet memes are a pervasive form of multimodal online communication; however, such communication often involves users from diverse linguistic and cultural backgrounds. Therefore, adapting memes across cultures and languages is a central challenge for enabling mutual understanding in online communication. Unlike ordinary translation or standalone text rewriting, cross-cultural meme transcreation must jointly preserve communicative intent, adapt culture-dependent meaning for the target audience, and maintain coherence between text and image. In this work, we first provide an explicit task analysis of cross-cultural meme transcreation and identify three core challenges: culture-specific knowledge understanding, intent and tone preservation, and multimodal consistency. Based on this analysis, we propose a multi-agent framework with specialized agents that are coordinated to address these challenges through cultural adaptation, target text rewriting, revision, and conditional visual adjustment. The framework strengthens target text adaptation with coordinated feedback to handle difficult cases that require deeper cultural or visual intervention. We evaluate the framework on bidirectional Chinese-English meme transcreation using both human evaluation and LLM-as-a-Judge. Our method consistently outperforms all baselines across both evaluation settings. In human evaluation, it achieves the best performance on all four dimensions and delivers a 33.1% average improvement over the strongest baseline, while in LLM-as-a-Judge, it attains the highest Top-1 ranking rate (60% versus 26% for the second-best baseline). Further analysis indicates that each component contributes to the performance. Our error analysis suggests that the remaining bottlenecks lie in humor reconstruction and image-text alignment rather than simple cultural knowledge gaps, pointing to future work on humor transfer.

Aesthetics are an important part of the symbolism of artistic works. Although subjective, humans categorize art based on the emotion evoked regardless of modality. What remains under-explored is how AI models form their own aesthetic categorization of human-produced media without explicit labels or cross-modal supervision. We present a self-supervised framework that projects four modalities (text, audio, image and video) into a shared 256-dimensional embedding space and applies iterative clustering to discover aesthetic structure. We discuss the divergence between AI-generated cluster assignments and human affective register labels on a weakly supervised multimodal dataset. This work has applications in understanding how AI structures cross-modal similarity, organizing heterogeneous media collections for Retrieval-Augmented Generation (RAG), and automated data labeling.

Reliable medical conversational AI requires authentic expert--patient interaction data, yet such datasets remain scarce, especially for low-resource languages such as Bengali. We present DocTalkBN, a large-scale multimodal dataset of real-world expert telemedicine conversations in Bengali, collected from nationally broadcast telemedicine programs featuring board-certified physicians. DocTalkBN contains 557.63 hours of paired audio and text, 1,515 multi-turn patient calls, 10,274 host--doctor question--answer exchanges, totaling 1.7M tokens, spanning 26 medical specialties. Unlike prior resources derived from medical forums, written health content, or synthetic data, our dataset preserves the spontaneity, contextual richness, and spoken characteristics of authentic medical interactions in a low-resource setting. To support benchmark-driven research, we further construct three downstream tasks from the corpus, medical triage classification, advice safety evaluation, and medical named entity recognition, and benchmark a diverse set of large language models and encoder-based baselines. Our results show that DocTalkBN is a practically useful resource, particularly for clinically grounded reasoning tasks. We release this resource to facilitate future research on reliable medical NLP and safer, more culturally grounded healthcare systems for low-resource languages. Our source codes and dataset are publicly available at https://anonymous.4open.science/r/doctalk.

Ontology learning from text remains challenging despite significant progress in Large Language Models (LLMs), which can hallucinate domain terms, produce inconsistent formats, and favor hierarchical over associative relations. In the LLMs4OL 2026 Challenge, we address both the End-to-End Flagship Task (Task A) and Ontology Extension Reuse Task (Task B) using an offline retrieval-augmented few-shot prompting pipeline. Our system employs Qwen2.5-14B-Instruct with all-MiniLM-L6-v2 for demonstration retrieval, selecting the top-5 examples for Task A and top-2 for Task B. A left-truncated context-windowing strategy preserves task instructions within long prompts. For Task B, generated triples undergo deterministic vocabulary-constrained filtering, retaining triples when at least one endpoint belongs to the sample's closed term/type vocabulary and removing duplicates of the initial ontology. The approach achieves Semantic Graph Similarity of 0.8692, Term-Typing F1 of 0.9200, and Taxonomy Discovery F1 of 0.8540 on Task B, while Task A achieves 0.7416 Semantic Graph Similarity. However, no non-taxonomic relations are extracted, highlighting limitations of closed, taxonomy-oriented relation vocabularies.

While most phenotyping platforms rely primarily on image-based measurements, advanced plant characterization requires the integration of active physiological sensing modali- ties such as chlorophyll fluorescence. We present an autonomous robotic platform designed to perform targeted fluorescence measurements on plant leaves. The system combines 3D plant reconstruction, geometric analysis, and motion planning to localize suitable measurement points and generate collision-free trajectories for a robotic manipulator. A dense 3D model of the plant is reconstructed from multi-view data and used to extract candidate leaf surfaces based on orientation, accessibility, and sensing constraints. These targets are then integrated into a task-level planning framework that guides the end-effector to precise contact or near-contact configurations required for point-based fluorescence acquisition. The platform enables automated, repeatable, and spatially resolved physiological measurements that go beyond passive imaging. By tightly coupling perception, geometric reasoning, and manipulation, the proposed system provides a robotics-driven approach to high-resolution plant phenotyping and opens new directions for autonomous agricultural inspection and plant-aware manipulation.

Enterprise AI deployment is a coordination problem across business units, application and AI teams, testing, platform engineering, infrastructure, security, operations, and data governance. Use-case benchmarks show whether one agent completes one task, but not how changing capabilities, models, runtime mechanisms, capacity, and enterprise data should be owned, changed, admitted, or evidenced together. We present four responsibility objects as shared organizational contracts: Skill (reusable, versioned capability and workflow asset), Harness (runtime compiler and governor), Scaffold (execution/control boundary and NFR owner), and a stack-external data substrate under independent CIO-governed semantics and telemetry. The runtime core is A = <S, H, X>, with the data substrate outside that stack. The central contribution is one bounded, falsifiable hypothesis, P1 (cost-aware capability-capacity separability): within a declared operating region, changing activated capability preserves the capacity-response interaction within a preregistered equivalence margin, while changing compatible Scaffold capacity preserves capability semantics up to a non-inferiority margin, and the required controls stay within a declared enforcement budget. Six design conditions become measured obligations whose coverage, violations, uncertainty, cost, and exclusions determine whether P1 is decidable. We propose a cluster-period randomized crossover experiment (balanced order, reset/washout, repeated seeds and failure regimes, cluster-aware uncertainty) with a four-state verdict: supported, falsified, conditional-engineering, or inconclusive. This paper contributes a contract-bounded runtime architecture, a source-preserving data substrate, and a falsifiable measurement protocol. It reports no completed implementation, experiment, dataset, or measured result.

Pretrained vision-language-action (VLA) policies provide strong priors for robot manipulation, yet adapting them online to fine-grained biomedical tasks remains challenging. Task success often hinges on subtle, view-dependent visual cues, while task-level rewards provide little guidance about which regions matter, making it difficult to learn task-relevant visual grounding from limited real-robot interaction. Online adaptation is further constrained by the computational cost of VLA inference and replay-based updates. We introduce GRAFT (Grounded Reinforcement Adaptation for Fast Task Learning), a framework for efficient online VLA adaptation through grounded perception. GRAFT uses region-level supervision to learn view-specific visual anchors that focus perception on task-relevant local cues without requiring region proposals at deployment. It further combines single-step action generation with cached visual-language prefix reuse to accelerate online learning. Across four biomedical manipulation tasks, GRAFT improves success rates by 25 percentage points under matched adaptation budgets, while reducing the computational overhead of online policy updates.

Explorable image-to-scene generation is essential for applications in gaming, robotics, and virtual reality. Existing methods based on video diffusion model (VDM) commonly rely on incomplete conditioning signals such as sparse point clouds or 2D panoramas, leading to stochastic hallucinations, long-term drifts and suboptimal 3D consistency. We present SpatialCrafter, a novel two-stage framework that addresses these issues by introducing a global 3D proxy for high-fidelity image-to-scene generation. Specifically, we decompose the generation process into global proxy generation and appearance refinement. For proxy generation, we propose a Point-anchored Sparse Structure~(PaSS) Flow module that predicts a spatially aligned and geometrically consistent 3D proxy. For appearance refinement, we re-frame the VDM as a Generative Deferred Refiner which synthesizes high-frequency photorealistic details upon proxy-defined scene geometry. To better integrate the proxy with the pre-trained VDM, we introduce Parallel Geometry Injection and Proxy-Aware Corruption training strategies, which improve robustness to proxy artifacts without disrupting the pretrained generative manifold. Furthermore, as no suitable dataset exists for this explorable scene generation task, we construct a new large-scale dataset of 115K scenes. To the best of our knowledge, it is the first hybrid dataset for image-to-scene generation. Extensive experiments on both synthetic and real-world datasets show that SpatialCrafter outperforms state-of-the-art methods, mitigates long-term drift, and remains robust and consistent under rapid camera motion and extreme viewpoint changes. Code, models, and the newly constructed dataset will be publicly released. See more at https://fangchuan.github.io/SpatialCrafter/.

Privacy-Enhancing Technologies (PETs) in computer vision often rely on noise or image perturbations to protect visual data while securely processing it, creating a trade-off between task performance and protection. This trade-off is commonly evaluated using image classification, which primarily captures semantic separability and remains robust despite significant geometric, spatial layout or local boundary alterations. As a result, it is too simplistic as a proxy for generic vision tasks. Exhaustive downstream-task evaluation, however, is computationally expensive because models must often be trained for each PET transformation and parameter setting. We therefore propose a compute-aware multi-task protocol for evaluating PETs in model training. It combines lightweight proxy tasks that target complementary aspects of visual structure while remaining simple and fast to compute. Across irreversible privacy transformations, key-based block primitives, and learnable image encryption schemes, we demonstrate that PETs with similar classification accuracy can differ substantially on other tasks. The outcomes highlight the need for PET evaluation protocols that move beyond classification-only reporting.

On-policy self-distillation (OPSD) has recently emerged as an effective post-training paradigm that improves policy optimization through dense token-level supervision from a privileged self-teacher. Despite its promise, OPSD remains largely underexplored for Video Large Language Models (Video-LLMs). Existing methods typically construct privileged teachers by augmenting their context with additional information while keeping the primary input unchanged for both teacher and student. Video reasoning, however, offers a distinct source of privileged supervision within the primary input itself: long videos contain substantial temporal redundancy, and only a small subset of frames provides the evidence necessary to answer a question. Building on this observation, we present $\textbf{Video-OPSD}$, an OPSD framework that exploits privileged visual evidence for both self-teacher construction and knowledge transfer. First, our Evidence-Grounded Self-Teacher conditions the teacher exclusively on annotated evidence frames while the student continues to reason over the complete video. This focused visual input enables the teacher to provide more informative supervision. Second, our Evidence-Guided Token Optimization adaptively weights token-level distillation according to each reasoning token's reliance on privileged visual evidence, thereby emphasizing perceptually grounded reasoning. Experiments across video understanding and reasoning benchmarks show that $\textbf{Video-OPSD}$ consistently improves upon Standard OPSD across multiple backbones and achieves performance comparable to GRPO while requiring substantially less training time, establishing an effective and efficient post-training approach for Video-LLMs.

Detecting the fetal abdominal circumference standard plane in low-cost obstetric blind sweeps is a highly imbalanced frame-classification problem: positive frames account for under 3% of a sequence, form short contiguous segments, and are poorly handled by off-the-shelf ultrasound and vision foundation models. We propose AnatoProto, a lightweight sequence-level framework that adapts a frozen BiomedCLIP encoder to fetal blind sweeps through four components: (i) anatomy-weighted spatial pooling that uses nnU-Net abdominal-region probabilities as a spatial prior to reweight BiomedCLIP patch tokens, so frozen semantic features are aggregated onto anatomically meaningful regions; (ii) a within-case prototype loss that pulls each frame embedding toward the mean of positive frames of the same sweep, exploiting case-level structure unavailable at the frame level; (iii) a three-stage cascade refinement (frame->segment->case-level rejecter) that lifts the prediction unit from noisy frames to structurally-constrained segments; and (iv) a hybrid prediction head that jointly models per-frame stability and inter-frame boundary transitions to suppress boundary false positives. On the ACOUSLIC-AI benchmark, AnatoProto reaches a test F1 of 67.72, outperforming the strongest foundation-model baseline (FetalCLIP + PRS, F1 = 54.52) by +13.20 F1 and the strongest video temporal-action-detection baseline (TriDet + PRS) by +15.76 F1. A synergy study, backed by embedding geometry and paired-bootstrap confidence intervals, shows that the prototype loss and anatomy-weighted pooling are not additive: applied alone the prototype loss reduces recall by 12 points, but combined with anatomy-weighted pooling it increases recall by 6.5 points -- a sign-flip we trace to the accuracy of the within-case prototype.

Reinforcement Learning with Verifiable Rewards (RLVR) and other RL-style post-training paradigms have been used for aligning large language models (LLMs) with reasoning standards. The resulting recent Reasoning Language Models (RLMs) such as DeepSeek-R1, o3, and Kimi k1.5 show that such RL-style post-training ("RL-for-LLMs") can substantially improve chain-of-thought reasoning, long-horizon planning, and self-correction. However, the computational footprint of these systems is massive: state-of-the-art RLM training requires millions of GPU-hours and tightly coupled multi-model pipelines that stress modern hardware far beyond classical supervised LLM training. This makes RLM training as much a parallel and distributed systems problem as an algorithmic one. In this work, to facilitate developing RLMs that are simultaneously high-performance, scalable, and cost-effective, we first systematize the RL-for-LLM paradigm and provide a compute-centric analysis of prominent post-training algorithmic frameworks: Proximal Policy Optimization (PPO), Group Relative Policy Optimization (GRPO), as well as their variants. Second, we develop a taxonomy of intra- and inter-model parallelism strategies for RL-for-LLMs, covering both traditional techniques (data, tensor, pipeline, sequence, context, and expert parallelism) as well as novel forms of parallelism and optimization techniques for multi-model RLM training, for example disaggregated placement, stage fusion, hybrid parallelism, and asynchronous execution. We harness the work-depth model of parallel computing to make our taxonomy and its insights rigorous and portable. Finally, we analyze existing RLM frameworks and we distill practical guidelines and outline open research directions for building scalable, fast, and cost-effective RLMs.

Multimodal representation learning has been shifting from traditional two-tower architectures to large language model (LLM)-based embedders due to their strong instruction-following capabilities. Despite this progress, existing approaches primarily focus on language and image modalities, which also remain the dominant modalities for user-conditioned interactions in current embedders. In this paper, we propose the first Omni-Interactive Universal Embedder (OmniUE), which not only learns a unified embedding space across text, video, and audio by leveraging intermediate-layer representations from dedicated learnable tokens, but also supports omni-interactive querying, enabling users to provide inputs in the form of text, visual regions of interest, and audio spans. Within OmniUE, visual and audio segmenters process diverse user interactions and integrate them with an omni-LLM to produce user-conditioned any-to-any embeddings via context aggregation. To evaluate OmniUE's omni-interactive capabilities, we introduce OmniCHOIR, benchmarking models for omni-interactive compositional audio retrieval based on the given text, video, and audio as well as unimodal or multimodal interaction prompts. OmniUE consistently surpasses state-of-the-art baselines across diverse modalities, with average improvements of 10.5% on textual-interactive video benchmarks (MMEB-v2-video), 1.1% on audio tasks (MAEB), 83.7% on visual-interactive benchmarks (SCaR), and 24.1% on our omni-interactive OmniCHOIR benchmark. We believe that jointly advancing omni-modal representation learning and omni-interactive querying paves the way toward universal embedders.

Accurately forecasting the movement of people in complex scenes requires reasoning over the past and present state of the entire environment. In this context, effectively incorporating object information and social interactions into a unified framework remains particularly challenging. To address this, we propose Object-Conditioned Social Diffusion (OCSD), a conditional diffusion model that integrates motion history, multi-person interactions, and object cues into a single framework. OCSD uses an object-conditioning mechanism that modulates denoising at every timestep, enabling fine-grained human-object reasoning, and a social encoder that models the interactions between all humans in the scene. As a result, our model naturally handles varying group sizes, complex social interactions, and supports sampling multiple plausible futures. Extensive experiments show that OCSD achieves state-of-the-art results on the Humans in Kitchens (HiK) and HOI-M3 benchmarks. It reduces the two-second path error by 121.5 mm (31.3%) on HiK and 130.5 mm (33.2%) on HOI-M3 compared to prior work, and produces more realistic long-term forecasts.

This paper proposes a fully differentiable jitter correction method for X-ray phase-contrast micro computed tomography using a deep learning-based image quality metric that estimates and compensates per-projection rigid jitter directly from the acquired projection data, without a pre-scan motion-free reference. The approach builds on a gradient-based auto-focus strategy adapted to parallel-beam geometry. A set of candidate objective functions is benchmarked in a controlled study, and the sensitivity of the visual information fidelity (VIF) metric to the jitter artifact is verified with the target phase-contrast data. To operate without a clean reference, a compact 3D convolutional neural network is trained to predict the VIF score from a single corrupted volume. A spatially selective total variation penalty applied exclusively to the image background is introduced to penalize spurious high-frequency structures that otherwise emerge during optimization. Experiments on biological specimens acquired at different synchrotron beamlines are conducted. Evaluation uses jitter motion applied to simulated and experimentally acquired projection data. The result confirms that the integrated pipeline reliably recovers fine structural detail lost due to jitter, with generalization demonstrated across morphologically distinct samples.

We introduce Riemann-1.0, a fully causal autoregressive World Action Model for embodied intelligence. Riemann-1.0 jointly models multi-view visual observations, robot states, and embodiment-specific actions within a unified causal autoregressive sequence, representing robot actions and world evolution as causal state transitions. Unlike existing WAMs based on joint generation, video-first prediction, or decoupled modeling paradigms, Riemann-1.0 unifies online robot policy execution and action-conditioned world simulation within a single model, enabling it to function as both an executable robot policy and a multi-embodiment visual world simulator. To scale embodied experience across heterogeneous data sources, we further develop a progressive embodied pretraining framework that unifies learning from egocentric human videos, handheld-gripper demonstrations, and heterogeneous robot trajectories under a shared World Action Modeling objective. Built upon 200K+ hours of interaction data, Riemann-1.0 progressively transfers large-scale embodied experience into executable robot manipulation capabilities. Riemann-1.0 achieves state-of-the-art performance across both simulation benchmarks and real-world manipulation tasks. It achieves success rates of 94.3% on RoboTwin2.0, 99.0% on LIBERO, and 62.6% on the long-horizon compositional benchmark RoboCasa-365, outperforming the previous best method by 8.4% On long-horizon real-world manipulation tasks, Riemann-1.0 achieves a Success Rate (SR) of 85.0% and a Progress Success Rate (PSR) of 94.4%, exceeding the strongest open-source baseline by 15% in SR. These results demonstrate that unified World Action Modeling together with progressive embodied pretraining effectively transforms large-scale embodied experience into generalizable robot manipulation capabilities.

Training molecular machine-learning models on ephemeral or memory-constrained accelerator instances can require repeatedly retrieving preprocessed molecular graphs from remote storage. FoldPipe is a lightweight Python orchestration layer for already-sharded PyTorch and PyTorch Geometric data. It retrieves one shard ahead in a background thread while the consumer trains on the current shard, keeping the number of live shard payloads bounded with respect to total dataset size. Asynchronous prefetch and bounded buffering are established systems techniques rather than novel scheduling algorithms. FoldPipe's contribution is a small integration targeted at native .pt molecular shards together with a source-pinned empirical characterization of its operating regime. We evaluate a SchNet energy-and-force workload on MD17 aspirin using 20 paired, order-alternating benchmark passes on a Tesla T4. Each pass processes five pinned shards containing 25,000 structures. FoldPipe records 16.33 s mean I/O-compute overlap, compared with zero by construction for the sequential bounded baseline. Mean pass time is 76.78 s for FoldPipe and 83.37 s for the baseline. However, the geometric mean paired speedup is $1.059\times$ with a 95% bootstrap interval from $0.878\times$ to $1.288\times$. The experiment therefore verifies the overlap mechanism but is inconclusive about a reliable wall-clock speed advantage under the observed public-network variability.

LLM-based agents are increasingly proposed for network fault diagnosis, but existing benchmarks evaluate them only on accurate tickets and always assume a fault is present, conditions rarely met in practice. We present FaulT-Bench, a benchmark of 200 troubleshooting scenarios across eight network topologies, five reimplemented from public practitioner labs, spanning genuine faults, false fault reports, incorrect device attribution, and incorrect root-cause claims. To isolate how ticket wording affects diagnosis, we further rewrite 72 false-premise tickets into five reporter personas that vary reporter confidence and verifiable detail one factor at a time, holding the network state fixed. Our automated harness deploys each scenario in Kathará, lets agents interact through the NIKA tool interface, and scores free-text diagnoses with an LLM judge across outcome, fix, and reasoning quality. Evaluating SADE, ReAct, and Claude Code, we find all three are near-saturated on accurate tickets and robust to misdirection, yet degrade sharply when the network is healthy and the ticket is wrong, probing until a benign condition can be promoted to a root cause rather than concluding nothing is wrong. Persona rewrites show that how a ticket is written matters more than what it claims: a confidently wrong report is handled about as well as an accurate one, while a vague, underspecified report degrades performance sharply. The three agents also fail differently, from constant over-diagnosis to unanswered runs, at very different cost. These results position FaulT-Bench as a benchmark for developing agentic systems that can reason reliably over the noisy, unreliable tickets of real-world network troubleshooting.

In the present work an efficient border control management procedure is proposed. Compared to operational queue management systems, whose operations are based on mostly static data, the proposed work takes into account dynamic traffic conditions, thus enabling optimal performance, even in cases of uncertainty. To this end, we are proposing a multi-modal Artificial Intelligence (AI) framework, tailored to th needs of border control systems, which enables real-time queue prediction, management, and resource optimization. The novel proposed approach integrates heterogeneous data sources and presents them through a unified representation by employing Long Short-Term Memory (LSTM) networks for queue forecasting. Furthermore, it leverages Model Predictive Control (MPC) and scheduling optimization to derive actionable control policies, which in turn can be presented to border control officers. The proposed work has been evaluated using synthetic data simulating realistic traffic. The evaluation results demonstrate that the proposed method reduces queue prediction error by up to 35% and average waiting time by 30%. Accordingly, the average throughput increases by nearly 20%, compared to ARIMA and rule-based methods. The abovementioned results show the effectiveness and efficiency of combining AI architectures with optimization techniques for proactive and adaptive border traffic management.

Reliable evaluation of vision-language models (VLMs) and medical vision-language models (Medical-VLMs) requires calibrated confidence, particularly under realistic clinical conditions. However, existing efforts mainly focused on improving accuracy, leaving calibration in the medical domain underexplored. To this end, we propose MVC-Bench, a calibration-centric benchmark for medical image classification with VLMs and Medical-VLMs. MVC-Bench assesses the calibration across three axes: (i) robustness to modality, backbone, and domain shift (ii) effectiveness of calibration strategies and prompt-tuning methods (iii) stability under prompt-template and random-seed variations. The benchmark covers eight different backbones, three medical modalities, including fundus imaging, histopathology, and chest X-ray under in-domain and domain shift settings. It compares post-hoc calibration, train-time calibration, and zero-shot inference methods, together with six prompt-tuning methods. Across more than 1638 controlled experiments, we report accuracy and Expected Calibration Error (ECE) as primary metrics, and further report results with complementary calibration measures, including Maximum Calibration Error (MCE) and Adaptive Calibration Error (ACE). We further investigate the underlying causes of miscalibration in VLMs and Medical-VLMs and propose a simple train-time calibration method, Multi-Class Margin (MCM) regularization, which achieves lowest ECE on 10 out of 12 settings in in-domain and remains competitive under domain shifts. Collectively, MVC-Bench provides a structured evaluation framework and actionable guidance for improving calibration in safety-critical medical workflows.

Representation learning has attracted great atten- tion and managed to reach good performances as a pretraining method for downstream tasks or as a first step towards unsu- pervised speech modeling. Yet, little is known about how such methods deal with out-of-domain speech and how could they be adapted in a few shot to new domains. This is important especially for accented speech where one observes a long tail of accents that diverge from the standard ones. We introduce ABX- Accent, a benchmark based on the AESRC dataset that features 10 different accents of English. It includes a small (< 10 hours) unlabelled training set in each of the accents and adaptations of the Zero Resources Challenge ABX evaluation metrics to each of the accents. We illustrate this benchmark with a baseline model that uses adaptive domain normalization to fine tune a pretrained Contrastive Predictive Coding model on the accents. This method is first developed on LibriSpeech using a male/female split. When applied to the new benchmark, the proposed method yields a relative improvement of 23.6% on across-speaker ABX scores on average compared to non adapted models. The data and metrics will be open sourced upon paper acceptance

Powerful code agents can execute scripts, call tools, and manage files, yet many important applications remain accessible primarily through graphical user interfaces. We argue that screenshot-and-click is an inefficient interface for software-operating agents: screenshots are state-incomplete, and GUI actions are brittle, semantically weak, and poorly matched to long-horizon planning. We introduce ASIL (Agent-Software Interaction Layer), an agent-native interface that exposes software through structured JSON observations and code-executable semantic actions, realized through the deepest feasible access path for each application. We instantiate ASIL across 15 applications and a benchmark of 300 single-application and 80 multi-application tasks. ASIL reaches above 80 with closed models while executing fewer than five actions per task. Under a repaired runtime and a 50-step screenshot budget, the same tasks yield 6.6 and 26.6 strict success under screenshot-and-click control, rising to 15.0 and 53.3 on an easier OSWorld-comparable band. Against application-native interfaces on matched tasks, ASIL exceeds LibreOffice's UNO API by 28-38 strict points but only matches draw.io's MCP content contract. The structured modality also suits training: small-scale SFT raises Qwen3.5-2B from 58.0 to 72.1 and Qwen3.5-9B from 66.6 to 80.4, and resource-limited on-policy RL further raises them to 74.4 and 82.2.

Large language models can summarize financial information, but an operational stock-research system must first assemble heterogeneous evidence, expose unavailable data and model capabilities, and control how generated opinions affect a final report. We present DSA, an evidence-aware orchestration framework for multi-market stock research with large language model (LLM) agents. DSA organizes the workflow into evidence acquisition, structured context construction, model-routed analysis, optional role and Strategy Skill reasoning, and report generation with selected context and diagnostics. A default report profile and an optional agentic profile share evidence and model-routing services but use profile-specific output validation and risk safeguards. In the agentic profile, core role outputs are processed by role-specific parsers, whereas Strategy Skill opinions undergo an additional signal-eligibility partition before synthesis; disagreement is supplied explicitly to the decision agent, followed by a conservative risk override. The reference implementation includes six regional market paths, fifteen bundled Strategy Skills, hosted and local model routes, and multiple execution and delivery surfaces. At a frozen software snapshot, a selected manifest of 1,457 portable offline backend contract tests passed; 596 cases were retrospectively mapped to six contract families central to the reported LLM-agent architecture. This evidence establishes implementation conformance for the tested software contracts, not superior report quality, forecasting accuracy, or investment returns.

Organizing long-term memory for multimodal agents remains challenging because existing methods either suffer from expensive question-agnostic offline summaries or naive embedding similarity matching that introduces incomplete and redundant context. To address these issues, we propose GraphMemix, a combinatorial-optimization graph memory framework that models memory organization as query-aware evidence-forest construction. Specifically, our method consists of three key components:(1) candidate graph construction, which expands multi-view seed memories through schema and semantic relations to acquire query-aware original context; (2) evidence utility and activation costs, which decouples direct memory support from anchor-conditioned relation verification to suppress redundant or conflicting information; and (3) forest optimization, which jointly selects a forest-format memory context under a maximum evidence budget and its reliable relational structure. By organizing memory into a query-relevant subgraph, the method avoids substantial lifecycle cost and recovers low-similarity complementary evidence. Experimental results across four long-term multimodal memory benchmarks demonstrate significant improvements with different foundation models and establish a new Pareto frontier between accuracy and lifecycle cost.

Large language model (LLM) judges are increasingly used across various evaluation scenarios, making their judgment capabilities valuable intellectual property. However, black-box access exposes these capabilities to model extraction attacks. Existing extraction methods do not specifically target LLM judges and provide limited support for multiple evaluation protocols under restricted query budgets. In this study, we propose JUDGESTEALER, the first query-efficient model extraction framework for replicating judging capabilities across pointwise scoring, pairwise comparison, and listwise ranking protocols. JUDGESTEALER exploits the strong cross-protocol agreement to acquire pointwise scores and transform them into pairwise and listwise supervisions without additional victim queries. To capture informative judge patterns and improve query efficiency, JUDGESTEALER dynamically selects pointwise inputs based on semantic diversity, predictive uncertainty, and potential judge biases. It further applies score smoothing and multi-protocol review to preserve the ordinal structure of scores and mitigate catastrophic forgetting during surrogate adaptation. Extensive experiments on state-of-the-art LLM-as-a-judge and reward models show that JUDGESTEALER consistently outperforms existing extraction baselines, achieving up to 73.3%, 87.0%, and 71.6% accuracy for pointwise, pairwise, and listwise evaluation, respectively. JUDGESTEALER also remains effective across different sur- rogate model scales, adaptation strategies, and reasoning settings. Moreover, JUDGESTEALER demonstrates robustness against representative extraction defenses.

Landscapes are named, but whether names retain measurable environmental information beyond broad geographic structure is rarely tested. We analysed 3,757 Welsh settlements using a frozen, source-audited 24-element lexical framework, preregistered outcome-specific models and geographically structured validation. The central comparison contrasted 101 settlements carrying high-terrain elements (\textit{bryn} or \textit{mynydd}) with 139 carrying low-terrain elements (\textit{cwm} or \textit{pant}). High-terrain names occupied locations 24.4 m higher relative to their 2-km surroundings (95\% CI, 10.8--38.1 m; Holm-adjusted $p$ = 0.00137). The association remained positive across prespecified 1-, 2- and 5-km neighbourhood definitions and was reproduced using an independently produced elevation source (24.1 m; 95\% CI, 10.6--37.6 m). Adding terrain-name polarity to a non-lexical spatial and settlement baseline reduced geographically held-out mean squared error by 4.63\%, 6.22\% and 7.30\% under 10-, 25- and 50-km spatial blocking, respectively, although improvement varied among held-out regions. River-related names provided weaker, directionally consistent evidence, while the preregistered woodland model was non-estimable. Residual spatial structure, unresolved name language and the absence of independent external replication limit interpretation. Selected Welsh settlement-name categories therefore retain measurable information about present-day terrain within Wales, without establishing individual etymology, causal naming, historical environmental memory or transferability to other naming systems.

Molecular crystal structure prediction (CSP) is important in pharmaceuticals, agrochemicals, and organic electronics, where subtle differences in molecular conformation and packing can strongly affect material properties. We present Packora, a flow-based generative model for molecular CSP that jointly predicts atomic coordinates and the lattice from molecular graphs. Packora supports multi-component and organometallic crystals and can condition on any subset of molecular conformers, stereochemical labels, and space-group information within a single model. Inspired by the CCDC CSP blind test, we evaluate generation and ranking separately, using generation to isolate generator quality and ranking to measure end-to-end performance under a common relaxation and ranking pipeline. We also systematically study architecture, training, conditioning, inference, and scaling, identifying an effective design based on cacheable pairwise reasoning, training objective and numerical solver choices, conditioning dropout, and balanced scaling of pairwise and single representations. Packora outperforms the baselines on both structure generation and ranking benchmarks, achieving the best matched-budget coverage across all six generation benchmarks, as well as higher experimental-form recovery, lower experimental-form ranks, and faster convergence in ranking.

Rehearsal-free class-incremental learning (CIL) with LoRA adapters remains challenging because the low-rank subspaces updated across tasks evolve without geometric control, causing unstable shared representations and repetitive collapse of task-specific updates into previously occupied directions. We introduce Geo-LoRA, a geometry-aware framework that explicitly regulates how low-rank subspaces, both shared and task-specific, evolve during continual learning. For the shared branch, Subspace Projection Preservation (SPP) constrains consecutive updates to follow smooth trajectories on the Grassmann manifold, and Adaptive Core-Slack Alignment (ACSA) decomposes transitions into principal and residual components, aligning the former while modulating the latter to balance stability and plasticity. For the task-specific branch, Median-Calibrated Block Overlap (MCBO) imposes a statistical constraint via normalized projection overlap, penalizing excessive reuse to mitigate subspace crowding. These constraints jointly regulate the evolution of all LoRA subspaces across layers and tasks without introducing additional adapter types beyond standard LoRA. Geo-LoRA provides a principled geometric formulation for continual low-rank adaptation and consistently achieves state-of-the-art performance across multiple benchmark datasets and different task lengths.

Reward models play an essential role in aligning visual generative models, yet most existing visual reward models use a single scalar score or rely on fixed criteria that cannot adapt to different instructions. This limits both interpretability and task sensitivity, especially for text-to-image generation and instruction-based image editing, where different inputs require different evaluation dimensions. We propose RubricRM, a pairwise generative reward modeling framework that first produces an input-specific rubric with evaluation dimensions, weights, and scoring criteria, and then applies the rubric to score candidate images. We train dedicated RubricRM models for text-to-image generation and image editing using a two-stage training pipeline: supervised fine-tuning teaches the model the rubric-based scoring paradigm, while GRPO further improves scoring through fine-grained dimension-level rewards. Experiments on multiple generation and editing benchmarks show that RubricRM outperforms existing specialized reward models and remains competitive with strong proprietary MLLM judges despite using smaller backbones. Our models, data, and code are available at https://github.com/zijiankan/RubricRM.

Feed-forward 3D Gaussian Splatting reconstructs an explicit Gaussian representation from multiple input images in one network execution, making 3D reconstruction increasingly accessible for casual captures. However, such captures frequently contain transient objects that appear in only a subset of the views. Such content can be encoded into the per-view Gaussians associated with the inputs that observe it and remain in the combined representation despite being observed by no other input. As a result, it may produce blurred, duplicated, or floating artifacts in novel views. We introduce a training-free filtering procedure that exploits this per-view prediction structure. For each input, we exclude its associated Gaussians and render the same camera using the remaining representation, revealing content that is inconsistent with the other inputs. Feature similarity forms candidate regions, and rendering-based verification retains only candidates whose removal reduces reconstruction error in the other input views. The procedure operates on a single frozen prediction without retraining or scene-specific optimization. Across three reconstruction models and two distractor benchmarks, it consistently improves novel-view quality with varying numbers of input views. On clean scenes, evaluations across four models show that the original reconstructions are largely preserved.

Large Language Models (LLMs) are evolving from performing end-to-end mathematical reasoning to integrating agentic intelligence. However, most existing math benchmarks evaluate only final answers. This outcome-oriented evaluation provides limited diagnostic value for identifying process-level failures or rigorous logic, failing to guide the transformation of LLMs into robust agents. To bridge this gap, we present a process-level benchmark designed to evaluate the inherent agentic mathematical reasoning abilities of LLMs. Our framework aligns problem-solving agentic behaviors with a structured taxonomy of reusable mathematical atomic capabilities. We design a comprehensive suite of planning, action, and feedback tasks across both textual and multimodal contexts, supported by an automated pipeline that synthesizes high-quality trajectories and produces fine-grained annotations via controlled LLM rewriting. Experiments reveal that models with similar end-to-end accuracy can exhibit markedly different agentic capability profiles. This demonstrates that process-level evaluation is crucial for interpreting the true potential of LLMs and guiding the development of next-generation mathematical agents.

Answering questions over real-world documents requires processing long inputs that interleave text with tables. Optical context compression, which represents context as images, promises to reduce token cost, but its effect on table understanding remains unclear. We study pixel-level table compression for question answering over documents with multiple tables, evaluating five VLMs across two benchmarks and five visual-token budgets. Representing tables as images at native resolution matches text in both performance and efficiency, but downscaling them makes models compensate the loss in readability with longer, less effective reasoning traces that cancel the expected savings. Highly downscaled tables, however, preserve enough signal to identify whether they are relevant to a question. We exploit this asymmetry with a training-free, two-step method: the model first identifies the tables needed to answer a question from a pixel-compressed context, and then reasons over those at native resolution. On long documents, our method saves 41% of total tokens and gains 7 accuracy points over single-step QA with native resolution tables. It also uses 15% fewer tokens than the most efficient single-step compressed configuration, with no accuracy loss.

Scene reconstruction with 3D Gaussian Splatting (3DGS) has become common, however deployment remains painful as the uncompressed file sizes can be massive. Current 3DGS compression systems combine multiple strategies for file size reduction, which can obscure where gains come from and limit component reuse across training pipelines. To make the gains more transparent, we propose KISS-GS, a modular compression pipeline named after the principle of keeping things simple, designed to decouple compression entirely from training. Given a 3DGS scene reconstructed with vanilla 3DGS, we are able to reduce it through compaction by 15.7x using a combination of state-of-the-art pruning schemes. Then we encode it into an image-based format designed for simple, ubiquitous decoding. With the SOG-XT format, we propose a novel extension to Self-Organizing Gaussians with two main contributions: (i) Self-organizing 2D Codebooks and (ii) Parallel Representative Assignment Smoothing (PRAS), which leverages the symmetry of quaternion and scale parameterizations to produce 2D attribute grids more amenable to encoding. This encoding reduces scene size by 6.6x. We show that optional encoding-aware fine-tuning yields a further 2.2x. Across standard 3DGS benchmarks, our simple and modular approach thus achieves a total of 85x to 319x reductions in the size of the scene over uncompressed vanilla 3DGS, setting new benchmarks for real-world scenes and surpassing tightly integrated methods in rate-distortion. Decoding relies solely on web-native image formats, and the modular design makes each stage easy to combine with future advances in reconstruction and compaction. Code and project page: https://fraunhoferhhi.github.io/KISS-GS/

Embodied agents need environments that are visually diverse, physically interactive, and changing over time. Procedural simulators can generate large interactive scene collections, and recent 4D generators produce compelling visual dynamics. Combining these properties in one environment, however, still demands extensive manual effort, and the result is rarely editable or controllable enough to reuse at scale. We present 4DSynth, a controllable procedural system that turns a natural-language description, a blueprint mask, or a single photograph into an editable 4D environment with explicit geometry, animated actors, collision-free trajectories, and physics-ready simulation state. Multiple scene routes share one geometry-grounded representation, so the same pipeline handles animation, camera planning, rendering, and task generation. To validate the full pipeline, we construct 4DSynth-Nav, an interactive navigation benchmark generated entirely from 4DSynth's procedural scenes. Two vision-language models evaluated across three difficulty tiers both fail the majority of tasks and stall after early subtasks. The same procedural controllability that produces these environments also makes each failure reproducible and each difficulty axis independently tunable. This paper presents both a controllable generation pipeline and the scalable benchmark it enables, offering a practical foundation for developing and evaluating embodied agents.

We present KinyaEmbed, the first dedicated sentence embedding model for Kinyarwanda, a morphologically rich Bantu language spoken by over 12 million people in Rwanda. Existing multilingual embedding models such as LaBSE, mE5-large, and OpenAI text-embedding-3-large perform poorly on Kinyarwanda due to severe under-representation in their pre-training corpora. KinyaEmbed is built on KinyaBERT-large and trained via a four-stage curriculum using MultipleNegativesRankingLoss (MNRL): Stage 1 leverages ~18,000 paraphrase pairs from the Official Gazette of Rwanda with three temperature scales; Stage 2 fine-tunes on 715 NLLB-translated MNLI triplets for entailment structure; Stage 3 aligns representations using English-Kinyarwanda OPUS-100 translation pairs; Stage 4 refines with 2,936 high-quality pairs filtered from KinyaCOMET at quality threshold 0.8. We evaluate on SemRel2024-rw and introduce Wiki-RW-STS, a new contamination-free Kinyarwanda STS benchmark of 300 pairs derived from Kinyarwanda Wikipedia. A seven-checkpoint ensemble (all5+23A*2, with the final stage double-weighted) achieves Spearman \r{ho}=0.7298 on SemRel2024-rw, surpassing mE5-large by 20.9% and OpenAI text-embedding-3-large by 41.0%. KinyaEmbed also achieves the best document clustering silhouette score (0.2146) across all evaluated models. All checkpoints, the KinyaCOMET filtered pairs, and the Wiki-RW-STS benchmark are publicly available.

Space-efficient warehouse layouts often contain single-agent-width aisles and dead-end workstations where robots have few places to wait without blocking others. In Multi-Agent Pickup and Delivery (MAPD) on such constrained layouts, robots must accept online pickup-delivery tasks while preserving protected waiting locations called Havens. The Safe HAven Retreat Planner (SHARP) introduced a mechanism that extends each committed task path with a validated retreat to the agent's dedicated initial Haven, but fixed-Haven commitments can send agents toward distant Havens after deliveries. We present A-sharp (Adaptive SHARP), which changes an agent's retreat target at task assignment time. A naive switch can cause two agents to rely on the same waiting location or let another committed path pass through a location that is still occupied or reserved. A-sharp prevents these failures with an availability test for candidate Havens and a pending-release rule that keeps the previous Haven protected until the agent departs. Under explicit Haven-structure and Safe Interval Path Planning (SIPP) assumptions, we prove invariant preservation and finite-release completeness: every task in any finite release sequence is delivered in finite time. Across 72,000 runs on 14,400 paired map-agent-count-rate-seed cases over four maps, both SHARP and A-sharp complete their respective 14,400 runs. For makespan (final delivery time), a prespecified paired comparison with Holm correction over all 138 configurations with more Havens than agents finds A-sharp significantly better in 107 configurations and never significantly worse than SHARP; on the tested tree map, the median reduction is 16.7%.

Accurate state estimation for autonomous underwater vehicles performing close-range seafloor sampling remains challenging. In low-altitude operation, down-looking cameras over featureless planar seabeds produce scale ambiguity, lateral degeneracy, and inconsistent feature tracking. Meanwhile, inertial-Doppler Velocity Log (DVL) fusion alone provides no mechanism for structural drift correction. We propose a Contact-Aided Factor-Graph Localization framework that treats physical interaction as an informative geometric constraint within a smoothing-based localization formulation. The method tightly fuses suction-based manipulator contact events with adaptive visual odometry, learned object detections, and on-board sensors. Visual odometry relative-pose factors and landmark bearing-range factors are uncertainty-scaled according to inlier statistics to prevent visually weak frames from destabilizing the estimator, while contact events are modeled as high-confidence factors that induce implicit loop closures without appearance-based place recognition. Furthermore, the system can fully initialize online during motion. Experimental evaluation in tanks, harbor, and simulation environments demonstrates that contact-induced constraints significantly reduce trajectory drift and improve object revisit accuracy compared to filtering-based navigation and contact-free graph formulations. These results highlight the role of embodied physical interaction as a localization primitive in perception-degraded underwater environments

Small language models (sLLMs) are nowadays hosted on devices with limited memory and computational budget. In an autoregressive setup, inference is memory-bandwidth bound: uniform quantization is often detrimental to such models, since their architecture has limited redundancies and only a few layers are not very sensitive to lower precision. We propose a composite metric that combines two orthogonal criteria: information retention (measured in terms of a normalized SQNR-based coefficient) and throughput gains (modeled using a roofline-based latency analysis). By profiling Gemma 3 1B, we find that Feed-Forward Network blocks and the embedding matrix are the most promising targets for acceleration. For each candidate, we estimate a normalized quality score based on simulated quantization and a normalized speed score based on roofline modeling with no actual execution needed. We combine the two scores in a composite priority coefficient, allowing us to tune the trade-off between speed and quality as needed. Our metric is general and can be used to prioritize individual blocks, their projection sublayers, or transformer layers as a whole. We evaluate our approach on several model architectures, showing that our estimates have at around 4% prediction error for the accelerated speedup. We find that our method generally allocates more resources to the most expressive layers compared to evolutionary search, specialized accelerators, or Shapley-value-based approaches that require expensive approximate inference. Our analytical approach makes sLLM quantization a predictable engineering task.

We present TabuLM, the first language model pre-trained on Kinyarwanda tabular data. Kinyarwanda is a morphologically rich Bantu language spoken by over 12 million people in Rwanda, yet lacks any dedicated tabular representation learning resource. TabuLM extends KinyaBERT-large, a two-tier morphological transformer, with additive row, column, and cell-type embeddings and a learned table-structure attention bias that sharpens same-row and same-column attention. Pre-training uses two new objectives: Masked Cell Recovery (MCR), which masks entire cells and forces reconstruction from row and column context, and Column Type Prediction (CTP), which predicts column semantic types from observed cell values. We pre-train on 172 Rwandan government tables (~35,000 cells) from NISR, RAB, REB, and MoH open-data portals, and introduce TabQA-kin, the first native Kinyarwanda table question-answering benchmark comprising 526 QA pairs across 31 tables and four question types. TabuLM achieves 62.0% exact match on TabQA-kin, outperforming KinyaBERT-large by 5.7 EM points and all multilingual baselines (mBERT 49.3%, XLM-R 50.0%) by 11.7-12.7 points. Analysis shows that structural table embeddings are most decisive for comparison and lookup questions, while morphological awareness provides complementary gains. Our code, data, and pre-trained checkpoint are publicly available.

Streaming autoregressive video models generate long videos chunk by chunk, using historical memory to maintain consistency. Existing methods typically expose subject and scene queries to history through similar policies. This stabilizes the subject, but can also lock backgrounds, viewpoints, and scene structure to previously generated states even when local motion continues. We call this failure memory-anchored scene under-progression; consistency and motion metrics alone can miss it. We introduce TetherMem, a training-free, query-aware spatiotemporal memory router for frozen video generators. TetherMem separates subject and scene queries and modulates historical access with region- and age-conditioned priors: subject queries retain identity-bearing history, while scene queries reduce reliance on subject history and stale backgrounds. Across 2,400 blinded pairwise judgments from 10 annotators, TetherMem achieves the highest estimated expected preference among eight streaming long-video baselines for overall quality (0.780) and scene progression (0.769). On complete 30-second videos, it sustains changes in background, viewpoint, and scene state while preserving subject recognizability and temporal continuity.

Automated candidate-job matching systems are increasingly classified as high-risk AI under emerging regulation, yet auditing them for demographic bias is expensive: classical correspondence-audit studies require hand-crafted resumes and manual submission, which does not scale to fast pipeline retraining cycles. This paper presents a general, reusable methodology that (1) uses task-specialized LLM agents to synthesize identity-neutral base resumes and inject controlled demographic treatments across five protected-characteristic axes (sex/gender, age, residence, language, disability), producing a K x (1+N) correspondence-audit matrix; (2) qualitatively flags inferred protected characteristics per an EU AI Act-aligned prompt; (3) ranks candidates against a job description via a fine-tuned sentence-embedding model and cosine similarity; and (4) computes a nine-metric fairness suite spanning counterfactual (score delta, mean absolute rank change, flip rate), group-fairness (top-K retention, four-fifths/impact ratio), and merit-aware (Recall@K, nDCG@K, equal opportunity, equalized odds) families, each with bootstrap confidence intervals, significance tests, and Benjamini-Hochberg correction, culminating in an automated PASS/INVESTIGATE/FAIL report with a composite risk score. On an example corpus of 5 job orders, 100 base candidates, and 10 demographic treatments (90 metric x variant evaluations): score shifts, top-K retention, and merit-aware rate gaps stay within tolerance for every treatment, but a rank-stability metric (MARC) and nDCG@K each surface borderline findings - including one on the neutral baseline itself - that a score- or retention-only view would miss. The results argue for multi-metric, multi-family auditing over any single aggregate score, and for LLM-agent-generated audits as a practical, low-cost complement to human-curated audits for any candidate-job matching pipeline.

As electricity market participants increasingly adopt learning-based agents for their bidding strategies, electricity markets are becoming algorithmic. Evidence from algorithmic markets in other domains shows that tacit collusion can arise purely through independent learning. Moreover, electricity markets are typically oligopolistic and feature repeated interaction among a small number of participants, making them structurally susceptible to non-competitive behavior. In the face of these observations, this paper investigates the hypothesis that tacit collusion may emerge in electricity markets where participants' actions are controlled by autonomous learning-based algorithms. We model strategic bidding as a repeated game with imperfect public monitoring, and model the participants' emergent behavior using multi-agent reinforcement learning. We propose a multi-dimensional set of criteria (going beyond profit comparisons against Nash equilibria) to assess whether the resulting behavior constitutes tacit collusion. Our experimental results showcase that such a danger is realistic for electricity markets: there are cases where agents do learn to sustain supra-competitive outcomes that are supportive of tacit collusion indicators, even though the agents were never instructed to collude.

Agentic recommender systems ground each decision of a large language model (LLM) in a persistent memory of the user, and in existing agents that memory is text: a narrative written and maintained by further LLM calls. Text limits this memory in two ways. It is updated one rewrite at a time, so exploiting the full interaction history is prohibitively expensive; and collaborative evidence, graded similarity over an entire catalog, does not survive translation into sentences. We propose CoVeMem (Collaborative Vector Memory), which vectorizes the collaborative core of the agent's memory. Frozen LightGCN user and item states form the memory bank; at each decision, the candidate set itself retrieves the most relevant historical states, which enter the LLM's context as soft tokens alongside a light textual profile. Contrastive alignment to item-semantic anchors, followed by listwise co-training with masked candidates, teaches the model to read these states and to rank through them; a pointwise yes/no readout scores each candidate. Across four instruction-grounded recommendation benchmarks, CoVeMem matches or exceeds the strongest collaborative text-memory agent on 19 of 20 metric cells while requiring zero additional LLM calls for memory maintenance beyond the shared static profile, against per-interaction calls for text memory. The memory now takes gradients: the full interaction history, out of reach for text, becomes available as training data for what the agent remembers and for how it reads what it remembers.

Robotic grippers face substantial challenges in grasping and manipulating thin objects. Most existing grippers rely on highly precise approach and grasp motions, which limits robustness and reduces applicability. This paper explores thin-object grasping using books as a representative example. Here, we propose a novel solution that integrates an active surface with underactuated compliance to achieve stable grasping of thin objects without complex control. First, an underactuated gripper with an active surface is designed. The active-surface thumb performs in-hand repositioning of the target book without requiring adjustments of the robot arm or the other fingers, while the underactuated fingers establish compliant contact conditions with the environment, and the reconfigurable structure enables reliable grasping of books under different configurations. Second, we establish a kinematic model of the gripper, and determine the initial grasp postures for two representative scenarios (books lying flat on a desktop and books vertically packed in a shelf). Third, by analyzing the physical model of a book lying on a table and its interaction with the gripper and the environment, we systematically optimize the structural parameters and grasping strategy. Finally, extensive experiments validate the effectiveness of the proposed gripper and strategy. The results demonstrate strong robustness and adaptability when grasping thin objects placed flat (including books, paper, fabric, plastic film, and mouse pad), as well as a high success rate when grasping vertically packed books. Moreover, the proposed gripper can reliably complete long sequential "grasp-place" tasks.

Industrial control systems (ICSs) rely on programmable logic controllers (PLCs) to connect networked computation with physical control. Tool-using large language model (LLM) agents represent an emerging attack threat: can an autonomous agent convert a network-reachable PLC into sustained adverse physical impact? However, existing evaluations focus on digital tasks or individual stages of PLC testing. In ICSs, evaluations that stop at software exploitation, an accepted write, or tool access may therefore mischaracterize physical risk. We present PLCBENCH, to our knowledge, the first real-PLC hardware-in-the-loop (HIL) framework for characterizing this cyber-to-physical capability and its boundaries. It combines vendor-native interaction, commercial PLC execution, closed-loop reduced-order process simulation, and independent outcome verification. A deterministic evaluator applies fixed rules to runner, communication, PLC-object, and process records to assign six hidden diagnostic flags, distinguishing usable PLC interaction, process-linked manipulation, and sustained physical impact. We instantiate PLCBENCH on four commercial PLCs crossed with four closed-loop workloads. Across five LLM families and 240 real-PLC episodes, 75 episodes (31.3%) sustain their respective physical objectives. Stagewise results show that 98 episodes stop before a valid native read, whereas 62 reach a process-linked write but do not sustain the final objective. Notably, richer process observation is associated with an increase in conditional objective attainment after a process-linked write from 44.2% to 64.0%. These measurements localize failure in configured PLC-process deployments and identify intervention points for future defense evaluation. To support reproducibility, we release the safely disclosable PLCBENCH code and a software-only reproduction pipeline through the accompanying artifact.

Fusing multiple modalities is expected to improve model performance. However, on the MultiHuSE dataset, early, late, and symmetric attention fusion often fail to outperform the best unimodal baseline (text). Pathway isolation of a symmetric attention fusion model reveals that the text-pathway accuracy drops from 74.9% to 56.4% after fusion in one such setting, indicating that the dominant modality can be degraded during integration. We term this strong-modality collapse and argue that it helps explain why some multimodal models fail to surpass unimodal baselines. We propose Inverted Asymmetric Fusion (IAF), which avoids forcing mutual attention across modalities. The dominant modality is preserved by passing through fusion unchanged, while weaker modalities attend to it as a contextual anchor. Before fusion, weaker modalities are strengthened using Modality-Aware Knowledge Distillation. We evaluate IAF on three benchmarks with different modality hierarchies: text-dominant datasets (MultiHuSE, UR-FUNNY) and an audio-visual-dominant dataset (MUStARD). Pathway isolation shows that IAF preserves the dominant modality's internal accuracy at its unimodal ceiling across all tested configurations, whereas symmetric fusion degrades it by up to 18.5% on MultiHuSE. IAF improves over the strongest unimodal baseline by up to 8.25%.

On-policy distillation (OPD), which leverages a pre-trained, specialized teacher model to provide dense supervisory signals, has achieved significant success in Large Language Models (LLMs) and has recently been adapted to flow matching models. However, this paradigm suffers from two major issues: First, training a separate, task-specific teacher for every new objective incurs high computational costs. Second, the discrepancy between teacher and student distributions often leads to compounding errors along the generation trajectory. In this paper, we introduce \textbf{Self-OPD}, a teacher-free OPD framework for flow matching models that turns the student's own self-exploration into step-wise supervision. At each timestep, Self-OPD branches the deterministic next-state prediction into $K$ stochastic SDE candidates, rolls them out with the ODE sampler, and compares their rewards against a deterministic self-reference baseline to obtain normalized advantages. The velocity field is optimized with an all-branch pull-push objective, where high-advantage branches attract the student and low-advantage branches repel it under direction-aware attenuation and SDE-variance normalization. For multi-objective alignment, Self-OPD fuses normalized scores at the reward level, avoiding direct gradient conflict. Experiments on single and mixed reward benchmarks show that Self-OPD outperforms prior RL and OPD methods without task-specific teachers.

Weak signals are early, low-visibility indicators that precede significant changes before those changes become established. Existing detection methods, based on keyword frequency, topic modeling, or untyped graph topology, fail to capture the semantic and relational structure through which such signals manifest. In this paper, we propose C-Unseen, a self-interpretable framework for weak signal detection in Dynamic Temporal Knowledge Graphs (DTKGs). We define a weak signal as a rare, semantically coherent subgraph that proliferates across consecutive TKG snapshots. The framework operates through two modules: a Rare Subgraphs Extractor, in which an LLM identifies subgraphs whose content is in tension with the dominant snapshot narrative via chain-of-thought reasoning, and a Weak Signal Alerter, in which the persistence of these rare subgraphs is tracked across time steps to isolate true weak signals. Experimental results demonstrate that C-Unseen outperforms keyword-, topic-, and graph-based baselines.

Recent advances in SLAM have leveraged 3DGS for photorealistic reconstruction and novel view synthesis. However, most methods rely on RGB-D input, which is unavailable on consumer-grade smartphones, and few integrate 3DGS within a collaborative framework. Therefore, we present CGS-SLAM, a hybrid decentralized/centralized system enabling multi-agent 3DGS SLAM using only RGB and inertial data. Each agent performs local tracking with inertial data as a motion prior and reconstructs a scaled map using a metric monocular depth estimator (Depth Pro). Keyframe encodings are shared among agents, enabling dynamic keyframing in regions of spatial overlaps with other agents, enhancing submap alignment. Afterwards, a central server aligns submaps using VGGT as a view alignment model. This bidirectional communication keeps communication cost low during mapping and global reconstruction in difficult GNSS-denied environments. Experiments on multiple datasets demonstrate competitive tracking performance, improved rendering quality over state-of-the-art methods, and accurate submap alignment.

Large language model agents reason, call tools, and act autonomously over many steps, but their agentic skills-correctly sequencing tools, planning under dependencies, judging untrusted inputs, and grounding generated arguments-are hard to measure with accuracy-only leaderboards. We present BekchiAI, which addresses both sides: a benchmark for measuring agentic skill and a platform for observing and controlling live agents. The BekchiAI-Benchmark, a suite of 13 tool-using ReAct agents across 7 task categories (arithmetic, structured/SQL, security detection, URL grounding, planning, orchestration, and tool-policy), totalling 2,057 deterministic, committed test tasks. Every task is verifier-checkable gold answers are computed by running canonical SQL against a real database, computing the exact schedule of a directed acyclic graph (DAG), or evaluating closed-form lambdas including adversarial security samples paired with deliberately imperfect signature scanners so a score reflects the model's own judgment, not the copying of an oracle. We define a small set of behavioral metrics beyond accuracy-tool-call adherence, URL hallucination and source-match, and per-model token cost and report a four-model comparison (Qwen3.7-Max, gemma-4-31B-it, gemma4:26b, gpt-oss-120b) whose story is in the per-family spread, not the aggregate. The benchmark runs are executed using the provided evaluation scripts. BekchiAI-Platform is a complementary web-based observability and control layer for deployed agents, providing full token and latency telemetry as well as remote run termination. The benchmark, evaluation tools, and platform are publicly released.

Many multimodal tasks depend on how visual elements are ordered and composed, not only on recognizing them in isolation. Internet memes are a compact case of this problem: their punchline often depends on a constrained reading order and cross-panel visual--textual cues. While large vision-language models (LVLMs) show strong performance on single-image understanding, it remains unclear whether they can perform sequence-aware reasoning over structured meme layouts, especially in Chinese social media. We introduce CMPM, a Chinese Multi-Panel Meme benchmark with 1,214 annotated samples covering five structural types, ordering dependency, panel-order constraints, and optional comment context. We formulate a two-layer evaluation: Task1 probes structure typing and order-sensitive panel sequencing (with a context ablation setting), and Task2 evaluates Chinese meme explanation generation with human ratings on five 1-3 Likert dimensions (visual, panel, humor, context, and faithfulness). We benchmark five representative LVLMs under a unified protocol. Results indicate that canonical-display accuracy is not by itself evidence of order understanding: the primary shuffled condition produces a sharp accuracy drop, revealing a persistent gap in order-sensitive multimodal reasoning. Task2 preferences place Gemini 3.1 Pro and GPT-5.5 above the open models, while comment context yields only a small and mixed Core4 gain. Code and data will be released upon acceptance.

Although Multimodal Large Language Models (MLLMs) have demonstrated impressive performance in Medical Visual Question Answering (Med-VQA), their reliance on global image features often lacks precise pixel-level grounding, thereby limiting clinical trustworthiness. To bridge the semantic gap between high-level clinical reasoning and spatial localization, we propose \textsc{\textsc{MedREAL}} (\textbf{Med}ical \textbf{RE}asoning-driven \textbf{A}nswering and \textbf{L}ocalization), a unified framework that seamlessly aligns linguistic reasoning with spatial grounding. Specifically, \textsc{MedREAL} introduces \textbf{S}eg \textbf{A}nchored \textbf{R}easoning \textbf{P}ooling (SARP) to distill task-relevant semantic evidence directly from \texttt{[SEG]} tokens within the MLLM's hidden states. Furthermore, a \textbf{R}easoning-to-\textbf{V}isual (R2V) fusion mechanism is proposed to effectively inject these reasoning-aware features into a segmentation pipeline for accurate mask decoding. To facilitate this paradigm, we construct MedRAVS-13K, a comprehensive dataset comprising 13,824 expertly validated samples across four diverse imaging modalities. Extensive experiments demonstrate that \textsc{MedREAL} significantly outperforms state-of-the-arts, achieving 68.49\% gIoU and 70.47\% cIoU on benchmark evaluations. By generating evidence masks that are strictly consistent with textual diagnoses, \textsc{MedREAL} provides a robust, interpretable framework for reasoning-driven medical image analysis.

User behavior simulation with large language models~(LLMs) is increasingly used to support multi-agent ecosystem simulation. Existing simulators typically rely on static user profiles inferred from historical observations, which become inadequate in socially intensive environments such as live streaming where interaction dynamics continuously reshape user behavior. We propose \textbf{LiveSim}, an LLM-based framework for live-stream ecosystem simulation. It represents users as editable behavioral hypotheses and progressively refines them through trajectory-grounded interactions, where discrepancies between simulated and observed trajectories reveal missing environmental shaping effects. These signals are further extracted as transferable environment-behavior patterns and accumulated in a collective behavioral memory to improve user-level behavioral fidelity and support ecosystem-level simulation. Experiments on real-world live-stream risk-control data validate the effectiveness of LiveSim in improving user-level behavioral fidelity and enabling ecosystem-level analysis of risk evolution and platform intervention effects.

Medical Visual Question Answering (Med-VQA) holds significant promise for clinical decision support, yet faces challenges due to limited annotated data and the high computational demands of existing large vision-language models. We propose MedFG-VQA, a lightweight framework that leverages a memory bank to augment DCT-based low-frequency features and employs graph-enhanced cross-attention for effective visual-textual alignment. Specifically, our approach features two key components: Frequency-Memory Fusion (FMF), which enhances low-frequency features by retrieving from a learnable memory bank built on DCT decomposition, and Graph-Aware Cross-Attention (GACA), which aligns visual-textual features via cross-attention and refines them through graph-convolutional aggregation. To address data scarcity, we construct SynMed-VQA, a large-scale synthetic dataset comprising over 2 million question-answer pairs across 9 imaging modalities and 10 major organs, generated with GPT-4o. Extensive experiments on SynMed-VQA and three other standard biomedical VQA benchmarks demonstrate that MedFG-VQA achieves competitive or superior performance compared to much larger models while maintaining significantly lower computational costs, highlighting its efficiency and potential for clinical deployment.

Interactive language-model agents use confidence signals to decide whether to answer immediately, retrieve additional evidence (from memory or external knowledge), or defer. Yet confidence is usually evaluated in isolation, without measuring the trajectory-level consequences of the actions it triggers. We propose matched trajectory replay, a controlled protocol for comparing confidence-to-action mappings. The protocol holds candidate answer states, evidence points, budgets, and action costs fixed. We use it to compare raw verbalized confidence with post-hoc isotonic calibration in a multi-hop question-answering system using Mistral, GPT, and Qwen models on HotpotQA and MuSiQue datasets. At the same numerical commitment threshold, calibration changes which questions agents ultimately commit to answering. Across all six model-dataset pairs, it increases accuracy among committed answers by up to 41 percentage points. However, it can reduce coverage and increase retrieval use. Overall accuracy improves by up to 15 percentage points on HotpotQA but falls by up to 17 percentage points on MuSiQue. These effects reflect a shift to a more selective, lower-risk operating point, not improved answers or confidence ranking. A calibration map fitted before retrieval improves held-out calibration through retrieval depths one and two, but is worse than raw confidence at depth three for all three models. Additional evidence helps on average, but this aggregate effect does not establish whether confidence identifies which individual episodes will benefit from another retrieval. Taken together, these results show that calibration can make commitment risk interpretable, but it does not estimate the expected benefit of another retrieval. Retrieval therefore requires a separate value-of-information or utility estimate. Evaluations should report held-out calibration, risk-coverage, and retrieval cost.

Large Language Models (LLMs) have demonstrated remarkable proficiency in natural language understanding, yet they struggle with strict multi-step reasoning, frequently suffering from hallucinations and inconsistency. Existing solutions like Chain-of-Thought (CoT) lack rigorous verification mechanisms, while standard Retrieval-Augmented Generation (RAG) often misses the complex, structural dependencies inherent in logical tasks. To bridge this gap, we propose a Neuro-Symbolic architecture that integrates a Logical Knowledge Graph (LKG) with dynamic solver routing. Specifically, we introduce an ontology-based LKG that treats logical rules and constraints as first-class topological nodes, enabling explicit modeling of dependencies extracted from text. We further design a Logic Router to dynamically dispatch tasks to the optimal symbolic engine, which is supported by a topology-aware hybrid retrieval mechanism. Experimental results on logical reasoning benchmarks demonstrate that our framework significantly outperforms state-of-the-art prompting and RAG baselines, delivering higher accuracy and verifiable reasoning paths.

Modern AI has greatly expanded the capabilities of image processing. However, the ready availability of powerful models, public datasets, and benchmark leaderboards has also en- couraged a model-first research pattern: researchers increasingly begin with an available architecture and optimize it on a public benchmark, rather than beginning with the underlying real-world imaging problem. This can produce impressive benchmark results without necessarily improving our understanding or solution of the real problem. This paper argues for a problem-first approach that distinguishes the physical imaging problem, solution principle, statistical estimator, and computational implementation, while clarifying what modern AI can achieve and which fundamental problems remain unsolved. Through case studies of super- resolution and low-light enhancement, we show how benchmark datasets may define tasks that differ substantially from the real-world problems they are intended to represent, and why performance improvements must be interpreted within the conditions under which they are obtained. We propose a six-stage workflow that places problem formulation, image acquisition, information-loss analysis, assumptions, ambiguity, and evaluation before model and dataset selection. The paper also proposes clearer standards for evidence, reproducibility, uncertainty, and claims of state-of-the-art performance. More fundamentally, it calls for a change in research culture and education so that future researchers learn to understand imaging problems deeply and use modern AI to achieve genuine scientific and technical advancement.

Large language models (LLMs) are increasingly applied to specialized domains, where effective use of domain expertise often requires reasoning over complex rules in concrete scenarios. However, existing benchmarks only partially evaluate this capability, as they either focus on output-level instruction constraints or overlook the distinct roles that rules play in scenario reasoning. To address these gaps, this paper introduces RuleWeaver, a benchmark construction framework for evaluating rule-centered scenario reasoning. RuleWeaver starts from corpus-derived IF-THEN Meta Rules, progressively augments them into complex rules, and composes these rules into rule-centered scenario QA instances. Beyond final-answer correctness, RuleWeaver further supports process-level evaluation through rubric-based answer quality, rule recall, and rule precision. Experiments on 11 representative LLMs show that current models still struggle with complex rule-centered scenario reasoning, with even the best-performing model achieving only around 50% of the maximum rubric score. We make our code and dataset available here: https://github.com/SharkSpicy-NLP/RuleWeaver.

Time series forecasting models operate on raw numerical sequences, lacking the semantic knowledge that domain experts implicitly leverage, such as the physical meaning of each variable, its statistical behavior, and its temporal dynamics. Recent efforts to bridge this gap fall into two camps. Some rely on large language models at inference time, which is computationally expensive. Others apply uniform textual prompts at the dataset level, ignoring the heterogeneous semantics across individual variates. We propose SAGE (Seeing and Augmenting with Grounded Encoding), an end-to-end CLIP-based framework that jointly models temporal, cross-variable, textual, and visual information. The CLIP text encoder processes frequency-enhanced patches and variable tokens, while gated residual paths inject variable-specific descriptions and statistical descriptors. In parallel, the frozen CLIP vision encoder aligns rendered series with temporal representations through a training-only contrastive objective. This dual use of CLIP adds complementary semantic and visual supervision without placing an LLM in the forecasting loop. Across eight long-term benchmarks and M4, SAGE achieves state-of-the-art accuracy. Ablations confirm complementary gains from multimodal alignment and variable-level knowledge.

Patient-conditioned acquisition policies for ECG lead-channel selection can outperform population-wide fixed protocols by tailoring the channel budget to each patient's observed cardiac state. However, the value of acquiring any given channel is defined relative to a downstream diagnostic evaluator, so marginal utilities learned under one evaluator need not transfer when the evaluator is replaced. We study this evaluator dependence empirically on PTB-XL by freezing two policies (ECG-on-Demand and MGA) trained with a controlled arbitrary-mask logistic evaluator, then scoring their unchanged acquisition trajectories with a more predictive masked raw-waveform ResNet1D. Exhaustive search provides metric-matched population-wide fixed comparators separately for each evaluator, enabling a clean interaction contrast. At budget $k=4$ on a held-out evaluation fold, ECG-on-Demand shifts from $D_\mathrm{C}=-0.011$ (favoring adaptive under the controlled evaluator) to $D_\mathrm{S}=+0.029$ (favoring fixed under the strong evaluator), yielding an NLL interaction of $+0.041$ (95% CI $[+0.030, +0.050]$). Across two policies, five budgets, and three probabilistic metrics, all 30 interaction estimates are positive with paired confidence intervals excluding zero. Three post-hoc sensitivity analyses -- common-reference scoring, training the strong evaluator on a mixture of policy-generated and random masks, and evaluator-aligned Strong-MGA policy training -- each preserve a positive interaction interval, making reference-choice and mask-distribution artifacts less plausible explanations. Evaluator-aligned training reduces but does not eliminate the gap. These results indicate that adaptive ECG channel allocation should be developed and validated jointly with its intended diagnostic backbone, and that jointly optimized sensing-diagnosis systems remain an open problem.

Vision-language-action (VLA) models leverage pretrained vision-language representations for robot control, yet simply adding historical frames does not reliably capture recent physical change. This is especially problematic in multi-stage manipulation, where visually similar states may require different actions depending on prior execution. To address this challenge, we present TemporalFlow-VLA, which learns compact execution history through physically grounded temporal supervision. Using recorded robot states, robot geometry, and calibrated cameras, we construct robot-surface temporal flow as a training-only target and supervise two execution-aligned temporal queries that provide structured history to the action expert. The geometric supervision path is not evaluated at deployment. TemporalFlow-VLA achieves 97.63 +/- 0.26% average success on LIBERO, including 96.60 +/- 0.87% on LIBERO Long, and 85.5%/84.2% Clean/Randomized success across 12 RoboTwin tasks. It shows its clearest advantage over prior methods on longer-horizon, multi-stage manipulation. Controlled history interventions show that action prediction depends on both historical content and temporal order. With asynchronous feature caching, temporal conditioning maintains single-frame-level server-side sampling latency without additional historical-encoding overhead. Overall, TemporalFlow-VLA provides a compact, physically grounded interface for exploiting ordered execution history without explicit motion estimation or geometric processing at deployment.

While generative AI has significantly advanced video editing, existing methods primarily focus on single-shot or short video clips. Editing long videos with multiple instructions remains a formidable challenge. Naive chunking strategies, e.g., fixed-duration segmentation, often lead to entity fragmentation, severe editing hallucinations, and disrupted temporal continuity. To bridge this gap, we introduce the Multi-Instruction Multi-Shot Long-Video Editing (MMLVE) task, which is structured around three core objectives: Cross-Shot Editing Consistency (CSEC), Multi-Instruction Decoupling (MID), and Zero-Destruction on Spatiotemporal Structure (ZDSS). To tackle these three unique challenges, we introduce an agentic editing framework that leverages the synergy of Large Language Models (LLMs) and Vision-Language Models (VLMs) to achieve shot-level video decoupling and precise instruction parsing. Furthermore, to comprehensively evaluate this task, we construct MMLVE-Bench, which is an MMLVE-focused dataset characterized by complex real-world spatiotemporal dynamics, high-density heterogeneous instructions, and sparse, random entity distributions. Three MMLVE-focused evaluation metrics are further exploited to assess the quality of the editing results. Extensive experiments demonstrate that our MMLVE-Agent outperforms existing closed-source SOTA approaches (e.g., Seedance 2.0), successfully eliminating editing hallucinations, preserving cross-shot editing consistency, and attaining seamless spatiotemporal transitions.

Travel planning agents assist users in generating personalized travel plans by modeling their individual preferences. Existing agents either rely on explicit user instructions or engage in multi-turn clarification to elicit user preferences. However, both approaches overlook the rich behavioral signals latent in users' past behaviors, which implicitly encode their preferences. This over-reliance on active user input increases interaction burden and limits plan personalization. To bridge this gap, we introduce a new task, Behavior-Aware Travel Planning, which infers user preferences directly from past behaviors and generates personalized travel plans. To facilitate research on this task, we introduce Behavior2Trip, a benchmark constructed from one of the largest Chinese online travel platforms, comprising 11,400 instances. Each instance represents an average of 39.8 past user behaviors spanning 14 attributes across 5 preference dimensions. We further propose B2T-Agent, a reinforcement learning-based agent that leverages user behavior trajectories, interacts with external tools for preference-aligned retrieval, and maintains an internal memory module. Experiments on Behavior2Trip show that GPT-4.1 achieves a full-constraint pass rate of only 0.5\% on the hardest tasks, while B2T-Agent built upon Qwen3-8B outperforms all baselines, highlighting the substantial challenge of this task. Moreover, Qwen3-8B trained with B2T-Agent also outperforms GPT-4.1 on the TravelPlanner benchmark, demonstrating strong generalization. Code and data are available at https://github.com/BUAA-IRIP-LLM/Behavior2Trip

With the growing demand for processing multiple image sequences in real-world applications, various visual token pruning methods have emerged to mitigate the computational and context length constraints faced by Large Vision Language Models (LVLMs). However, most existing pruning approaches rely on static strategies that struggle to adapt across different architectural LVLMs and multi-image scenarios, and are additionally constrained by their dependence on attention computations that are incompatible with efficient techniques like FlashAttention. To address these limitations, we propose a training-free, Adaptive Visual Token Pruning (AVTP) framework, applicable to diverse LVLM architectures. We strategically determine pruning layers based on empirical analysis of visual attention distributions across various LVLMs, and implement adaptive pruning ratios in multi-image contexts where images of higher importance retain proportionally more tokens. We conduct extensive experiments across different LVLMs to demonstrate the effectiveness and robustness of AVTP. Specifically, Qwen3VL-8B achieves 2 times inference speedup while maintaining 96.1\% of its original accuracy on multiple multi-image benchmarks, InternVL3.5-8B retains 94.1\% accuracy, and LLaVA-OV-7B even exceeds its original baseline performance. Our code is available at \href{https://github.com/zry13/AVTP}{this link}.

We present an online learning framework that enables a bimanual robot to acquire diverse juggling patterns directly on physical hardware within minutes, even with a significant sim2real gap. One of the most important lessons from this work is that a model, even when far from reality, can be extremely useful for learning. This motivates a central philosophy of our approach: learning should build upon the robot's current knowledge rather than replace it. Our regularized memory-based learning puts this principle into practice by learning a local model from accumulated experience while retaining the global prior model to extrapolate where experience is sparse. This enables efficient and stable online learning from each new experience without resorting to uninformed exploration over a vast space of possible behaviors. Equally important to continual on-robot learning is safety, allowing the robot to repeatedly practice and improve in the real world. We construct a mutually reachable set that allows safe transitions between successive throws and catches, without driving either arm into a state from which its next action would require violating the robot's joint or actuator limits. Together, these ideas enable a bimanual robot with multi-fingered hands and onboard vision to safely learn and compose five canonical three-ball juggling patterns, including cascade, tennis, half-shower, shower, and box, within less than 5 minutes of real-world interaction. More broadly, this work points toward robots that build upon imperfect prior knowledge and continually refine their behavior through their own real-world experience.

Scaling video generation to long durations reveals a critical bottleneck: current models lack robust long-term memory. This deficiency can be studied along two critical aspects: object permanence, the ability to precisely reproduce the appearance of objects upon re-entry; and memory capacity, the ability to process ultra-long context and use information from distant history. Robust long-term memory requires both: object permanence without sufficient context handling limits the temporal scope, while long context length without permanence fails to maintain identity. To address this, we present Ring Forcing, an autoregressive video diffusion framework designed to robustly construct and precisely utilize long-term memory. Our ring-structured training strategy enforces retrieval from distant history, effectively reconciling the trade-off between strict historical adherence and generative diversity. To expand memory capacity, we introduce a compression and timestep composition strategy. Under fixed sequence length constraints, this method extends the effective historical span to minutes-long durations and achieves a comprehensive receptive field over the entire history. Furthermore, we present a sparse RoPE mechanism to enable flexible, scalable memory adaptation while fully exploiting pre-trained priors. Extensive experiments demonstrate that Ring Forcing achieves superior minutes-long coherence and object permanence, significantly outperforming state-of-the-art methods.

Recent work shows that pre-trained, instruction-tuned vision-language models (VLMs) perform well at mapping from instructions and observations to high-level plans, but struggle to realize such plans as reliable low-latency action sequences in unfamiliar environments. At the same time, world-model controllers excel at fast observation-to-action control, but lack open-ended task guidance. In this work, we combine these strengths into a single system, Instruct-to-Act, where we train a world-model controller to act autonomously at high frequency when conditioned on sparse, higher-latency, and high-level text instructions generated by a VLM planner. To train controllers to be language-instructable, we relabel segments of controller policy rollouts with synthetic instructions and jointly optimize a behavior-cloning objective along with existing reward-maximizing and world-modeling objectives. We evaluate our proposed approach across seven embodied environments, including three multi-agent environments where VLM planners coordinate through language while trained controllers serve as their actuators. Under matched observation and action spaces, our decoupled approach consistently outperforms controller-only and direct VLM action-generation variants, preserves fast control, and lets us swap in different pretrained VLM planners without fine-tuning, while remaining competitive with strong vision-language-action and multi-agent RL baselines on six of seven tasks.

Control system design is critical for modern industry, such as chemical process temperature regulation and aero-engine control. However,traditional control design workflows rely heavily on expert knowledge and extensive manual parameter tuning, resulting in limited efficiency and scalability. To this end, this paper proposes AI Control Scientist (AICS), the first large language model (LLM)-driven agent capable of automatically generating optimized controller from language design requirements. Specifically, a Task Modeling Agent interprets user requirements to engineering constraints; a Controller Design Agent generate candidate controller structures and executable code; and a Parameter Tuning Agent refine controller parameters under closed-loop performance criteria. Experiments demonstrate that the proposed agentic system can automatically generate multiple representative control systems, outperforms existing automated baselines in both design success rate and optimization efficiency. This work has the potential to transform control system design from human-driven to agent-driven, paving the way for model predictive control and other advanced control systems design.

Preference learning optimizes models using response pairs, yet the informativeness of these pairs is fundamentally shaped by the instructions from which they are generated. We identify instruction quality as a hidden bottleneck in preference learning: low-quality or ambiguous instructions restrict the response-quality distribution, limiting strong chosen responses and weakening preference signals. Through Best- and Worst-of-N analyses, we show that instruction quality constrains both the ceiling and floor of sampled response quality. Motivated by this observation, we introduce an instruction-refinement pipeline that selects weak instructions using reward signals and revises them with rubric-guided LLM feedback, improving preference data without discarding examples. Across offline and online preference learning settings, experiments on multiple models and benchmarks show broad alignment improvements over original data and alternative data-improvement strategies. Further analyses indicate that instruction refinement raises achievable response quality and complements response-centric preference data curation. Overall, instruction quality emerges as a key factor governing how informative preference signals are formed for LLM alignment. Code is available at: https://github.com/01choco/instruction-refinement/

Action-conditioned cloth dynamics prediction requires both locally plausible deformation and long-range coordination. Existing approaches largely follow two paradigms. Mesh-based GNNs capture local physical responses through material connectivity. However, their finite message-passing range limits coordination between topologically distant regions, while autoregressive rollouts tend to accumulate prediction errors. Transformer-based dynamics models capture long-range interactions through global attention, but often operate without explicit material connectivity and must learn local topological responses directly from data. We propose MeshPriorDiT, a hierarchical dynamics model that decomposes future cloth motion into a structured mesh prior and a generative residual. An action-conditioned mesh GNN first predicts multi-step vertex displacements, yielding a reference trajectory that respects material topology and grasp constraints. Conditioned on historical states, planned actions, and the mesh prior, a Residual DiT then uses conditional flow matching to jointly generate the residual motion not captured by the prior. The generated residual is further rescaled and decoded using material adjacency to coordinate corrections across neighboring vertices. We evaluate MeshPriorDiT on 15-step autoregressive rollouts across three cloth manipulation tasks. Averaged over the three tasks, MeshPriorDiT reduces average Global MSE by 43.42% relative to the GNN-Only baseline and by 75.03% relative to the DiT-DDPM baseline, while maintaining a favorable Edge-strain MSE comparable to that of GNN-Only.

Dense warehouses often contain single-lane aisles, dead ends, and tree-like guidepaths that leave little room for idle agents to wait without blocking others. Existing Multi-Agent Pickup and Delivery (MAPD) guarantees for completing all finitely released tasks typically rely on extra waiting endpoints that planned paths can avoid, or on biconnected topology; these assumptions may fail in such layouts. We study fixed-Haven reservation for online MAPD, where pickup-delivery tasks are released over time. Each agent owns a fixed Safe Haven (Haven for short), usually its start cell, that only the owner may occupy and that other agents treat as blocked. For finite task releases, we prove that this fixed-Haven contract completes all released tasks under Haven-Reachability and explicit planning/progress assumptions. We implement the contract in SHARP, a Safe-Haven Retreat Planner that keeps every busy or retreating agent on a collision-free reserved route ending at its Haven. We compare SHARP with representative TP and PIBT-family MAPD baselines: Token Passing (TP), Priority Inheritance with Backtracking (PIBT), and PIBT with Temporary Priority and Temporary Avoidance (PIBTTP-TA) for biconnected main areas with attached trees. In the robustness sweep, SHARP is the only method with 100% success on all tested configurations, at substantially higher centralized planning cost on tree-like layouts. A TP-style fixed-home-return counterfactual with full-route validation also recovers robustness on tested tree-like layouts, suggesting that fixed return is a central robustness mechanism there. A no-overwrite variant shows that disabling mid-retreat reassignment worsens service time (release-to-delivery latency) by 1.89 times and makespan by 1.53 times in the tested high-load tree condition.

Faithful chart generation in real-world data-science workflows requires grounding visualizations in scattered evidence, computing chart-ready quantities, and rendering them accurately. Modern LLMs can produce visually plausible, instruction-compliant charts, yet data-level hallucinations remain difficult to detect in long, noisy, and multimodal contexts. To measure this gap, we introduce DEEPCHART, an expert-annotated benchmark of 1,482 task-conditioned chart-generation instances drawn from real-world scientific papers, financial filings, and ecosystem reports. DEEPCHART formulates chart generation as an Extract--Reason--Visualize pipeline and evaluates source-data extraction, derived-data reasoning, and chart rendering stage by stage. Experiments with state-of-the-art models show that visually plausible charts often conceal data-level hallucinations, with extraction and reasoning errors common in realistic long and multimodal settings. These findings suggest that larger context windows alone are insufficient; faithful chart generation also requires reliable evidence extraction and quantitative reasoning before rendering. Our benchmark and associated resources are available at https://github.com/tangdouer1005/DeepChart.

LLM agents used for scientific experimentation must do more than generate executable code: they must implement the reference method faithfully, design experiments that test the paper's claims, and provide evidence supporting those claims. We show that agents often produce methodological hallucinations: silently reducing datasets or training budgets, replacing failed learning or generative components with lookup or oracle functions, or drawing conclusions from resource-limited settings where a method's claimed advantage disappears. To detect these failures, we introduce ABE-Ralph, a reference-anchored auditing framework that represents claims, protocols, required components, baselines, and metrics as structured experimental constraints, guides implementation through an 8-step workflow, and performs quantitative, qualitative, and code-level verification. Across 30 long-horizon reproduction runs covering 12 machine learning domains, ABE-Ralph achieves a 93% robust execution rate and identifies five scientific failure modes. In 23 NatureBench discovery tasks, ABE-Ralph matches or exceeds state-of-the-art performance on 5 tasks. These results show that reliable evaluation of AI scientists must assess whether the experimental design faithfully tests the intended claim and whether the resulting evidence supports it, rather than treating code execution or plausible metrics as evidence of scientific success.

Glass surface detection (GSD) is critical for scene understanding and reconstruction, and yet remains challenging due to the transparency and reflectivity of glass surfaces. Existing GSD methods typically rely on 2D appearance cues, which may fail in geometrically ambiguous scenes. In this paper, we propose a paradigm shift: grounding GSD in 3D visual geometry to explicitly model the physical existence of glass surfaces. Our method first distills rich 3D priors from the visual geometry grounded transformer (VGGT) and generates glass-aware 3D representations. It then exploits multi-tasking learning with a novel glass detection head, consisting of two core modules: a Frequency Self-Attention Module (FSAM) that identifies glass-specific spectral features for glass surface localization, and a Geometry Grounding Block (GeGB) that selectively grounds 2D features in 3D geometry for glass surface segmentation. Extensive experiments demonstrate that our method achieves state-of-the-art performance across seven standard GSD benchmarks, generalizes well to video/multi-modal data, and substantially improves reconstruction in glass scenes. Code is available in https://github.com/YT3DVision/VGGT_GLASS.

Data lakes rely on metadata to remain usable, yet this meta data is often limited or weakly informative for column relationship discovery, especially in ERP-derived datasets with coded or abbreviated schema labels. We propose ColRel, a two-stage method that builds column embeddings from metadata and data available at ingestion time. In difficult cases, such as coded schemata, business dictionaries help better interpret column names and support the generation of short natural-language descriptions used in the second stage. Experiments on public benchmarks and an industrial ERP dataset show that ColRel is particularly effective in semantically related, weak-signal settings.

Scientific LLM agents have shown promise in literature reasoning, tool use, and experiment planning, but it remains unclear whether they can autonomously improve large, tightly coupled scientific machine-learning systems through executable code changes and computationally expensive validation. We study this question in protein folding, where progress requires coordinated architectural modifications, multi-objective evaluation, and domain-aware interpretation. We present AgentFold, a multi-agent framework that formulates folding-model development as a closed-loop search over executable code variants. Starting from ESMFold, AgentFold proposes hypotheses, implements and debugs code-level modifications, evaluates model variants, analyzes experimental outcomes, and stores both successful and failed interventions in structured memory. An MCTS-style policy allocates computational resources across high-scoring search branches. On an engineering-scale protein-folding codebase comprising more than 2,000 lines of code, AgentFold explores approximately 80 model variants using approximately 5,000 GPU-hours and 170 million LLM tokens. Under a matched computational budget, AgentFold improves the best lDDT by 7.5% over independent Codex proposals and outperforms a random-search control. Beyond model improvement, the resulting intervention traces reveal recurring empirical design patterns: stable gains tend to arise from early, soft, learnable priors and gated refinement, whereas direct geometric perturbations and geometry-conditioned feedback often destabilize training. The code and experimental resources are publicly available at https://github.com/lmqfly/AgentFold.

Zero-shot classification needs efficient label retrieval and fine-grained visual reasoning, yet discriminative and generative vision-language models fail in complementary ways.When CLIP's top-1 prediction is wrong, the correct label often remains in its top-$K$ shortlist, making disambiguation rather than recall the key challenge.Standalone generative models, however, are hindered by large label spaces and unconstrained outputs.This complementarity motivates separating broad candidate retrieval from fine-grained, image-grounded verification.We propose G2D, a training-free framework that uses a generative VLM to verify CLIP-retrieved candidates against the image.Candidate names and CLIP probabilities provide a structured prior for resolving visually similar classes.Fixed confidence routing, entropy-adaptive candidate sizing, and trie-constrained decoding focus generative reasoning on uncertain samples and ensure one valid output for each input at test time.Across eight benchmarks, G2D achieves 68.85% average accuracy, versus 59.35% for CLIP and 63.11% for the standalone VLM.Across seven generator configurations, candidate-set verification improves average accuracy by 1.08--27.42 percentage points.G2D also transfers to DCLIP, WaffleCLIP, and CuPL, supporting a practical interface between discriminative proposal and generative visual reasoning. Code: https://github.com/Harzva/G2D

Enterprises fine-tune language models on proprietary data that may later require removal due to privacy, contractual, or compliance obligations. Selective unlearning removes requested knowledge while preserving model utility, offering a practical alternative to full retraining, but existing methods treat the explicitly identified forget examples as the complete deletion scope. This is insufficient when target knowledge remains recoverable through paraphrases, aliases, or neighboring training examples. We propose GRAPHSU, a graph-guided controller that expands the deletion scope beyond forget seeds by constructing a weighted support-route graph, propagating deletion pressure through it, and applying graded forgetting strengths to high-risk neighbors. On the Task of Fictitious Unlearning (TOFU), a synthetic author-profile question-answering benchmark, and PISTOL, a structural-unlearning benchmark built around interconnected factual samples, with GPT-2 Medium and Llama-3.2-3B-Instruct, GRAPHSU achieves the lowest utility-feasible soft leakage across all deletion settings, reducing leakage by up to 49.5 percentage points over a matched seed-only baseline, demonstrating that effective enterprise unlearning requires controlling support routes, not just forget seeds.

Outdoor LiDAR semantic scene completion (SSC) recovers a dense semantic voxel grid from a scan observing 1% of the target volume, under class imbalance beyond 7,000x. We recast SSC as generative semantic scene completion (GSSC): a single discrete-diffusion formulation in three roles. First, paired sparse-dense scene synthesis (PS$^3$) generates matched sparse LiDAR observations with their dense semantic completions, addressing the long tail at its source and yielding the PS$^3$-SemanticKITTI corpus we train on alongside SemanticKITTI. Second, semantic-guided generative scene completion (SGSC) generates the scene from noise with multinomial discrete diffusion, conditioned on the sparse scan through a bird's-eye-view semantic map and a sparse 3D feature stream. Third, the same framework instead refines an existing completion in one flow-matching step: structured source discrete diffusion (S$^2$D$^2$). S$^2$D$^2$ improves the mIoU of SGSC's own output and every external SSC base tested, without base retraining or test-time adaptation. On the strongest base, one step without test-time augmentation reaches 38.8% mIoU on the SemanticKITTI hidden test. To our knowledge that is the best causal, single-sweep, single-sample result on that leaderboard, +2.1 pp over the previous best published score under the same restriction. Four correction steps with eight-view test-time augmentation reach 39.2%, outside that restriction.

Agent skills bundle instructions, reference data, and executable helpers that let a general agent perform specialized tasks. Hosted providers can keep these files secret while selling access to task results, making the skill itself a valuable target. Existing disclosure defenses can block requests that ask for the skill or reproduce its text, but they cannot block customers from submitting the ordinary tasks the service is built to complete. We present Daydreaming, an execution-only attack that steals a multi-file skill through black-box task interactions. The victim is never asked to reveal the skill or grade a reconstruction. Instead, Daydreaming adaptively creates crafted tasks whose results distinguish possible hidden behaviors. It tests individual behaviors, uses attacker-controlled shadow agents to choose a design, and completes each file using stored victim results and local execution checks. We formalize three nested threat levels of access as Differential, Trace, and Output, and focus on Output, where the attacker sees only the final response and returned files. Across 7 skills and 4 victim models, Daydreaming recovers 86.8% of the original skill's capability at Output, outperforming SigLeak by almost 4x. It produces installable skills using a median of 32 victim calls per skill even with disclosure defenses enabled. These results show that hiding skill files and filtering direct disclosure do not, by themselves, prevent functional reconstruction through normal use.

Graph neural networks (GNNs) are typically conceptualized as message-passing neural networks, yet it remains unclear why neighborhood aggregation reliably outperforms node-wise multilayer perceptrons (MLPs). Despite its empirical success, this paradigm can be computationally expensive and sensitive to imperfect graph structures. In this work, we present a retrieval-augmented view of GNNs: each layer makes predictions by applying an MLP to a node representation together with a permutation-invariant summary of retrieved graph context. Motivated by this perspective, we propose RTA, a simple MLP-based framework that replaces structural message passing with label-aware retrieval and propagation. We provide theoretical insights that (i) connect retrieval-based aggregation to softmax-attention message passing, and (ii) establish the robustness of retrieved-context supervision to mis-retrieved outliers. Experiments on multiple text-attributed graph benchmarks show that RTA matches or even outperforms strong GNN and graph LLM baselines while improving efficiency and robustness across diverse scenarios.

Large language models offer broad capabilities, but adapting them to evolving domains, tools, and requirements often entails repeated post-training. Autonomous systems automate parts of this process by proposing updates, training candidates, and using evaluation feedback to select subsequent proposals. As evidence accumulates, a central problem emerges: which past update evidence remains actionable after subsequent training has changed the parent model? An update's effect depends on its parent, data, and training stage. Treating past success as context-free permission can waste compute. If the resulting child is promoted, it can also degrade the subsequent training trajectory. We formulate this problem as conditional experience transfer and introduce Boundary-Calibrated Intervention Transfer (BCIT), a method that authorizes experience reuse before weight-changing training. BCIT binds an observed effect to its source context, checks applicability conditions, vetoes candidates with named hard conflicts, and obtains current-state evidence through a bounded training trial when needed. Fully trained candidates still face a shared adoption rule, and only observed events extend memory. On one 4B model adapted across finance reasoning, text-to-SQL, and function calling, candidate updates exhibit heterogeneous target and retention effects across the evaluated contexts. Under matched candidates, evidence, and compute, BCIT authorizes fewer harmful updates and attains higher equal-budget final-model quality than the evaluated alternatives. These results support treating experience authorization as a distinct problem in autonomous post-training.

In recent years, transductive knowledge graph embedding models have been applied to tasks such as link prediction and query answering. Although knowledge graphs often contain rich numerical attributes, most embedding models neglect them, limiting their ability to represent real-world knowledge graphs with diverse information. In this work, we propose a neural regression model (LitEm) that enables transductive knowledge graph embedding models to predict numerical attributes within knowledge graphs. Experimental results demonstrate that LitEm achieves the best or second-best results on most attributes across FB15K-237, YAGO15K, DB15K, and Mutagenesis. Furthermore, we propose a co-training framework that jointly trains state-of-the-art transductive knowledge graph embedding models with LitEm, which improves link prediction performance mainly for bilinear models and simultaneously enables them to predict numerical attributes. In addition, the literal-awareness evaluation demonstrates that co-training helps models to encode and exploit attribute information in a "literal-aware'' manner, suggesting that the observed gains are not merely due to additional parameters. We publicly release our implementation at https://github.com/dice-group/dice-embeddings.

Robotic and edge intelligence systems operate in dynamic environments where data arrives continuously, requiring models to adapt while preserving previously learned knowledge under strict memory and energy constraints. While parameter-efficient fine-tuning has shown promise for continual learning with vision transformers, conventional architectures rely on dense computation and remain costly for real-world deployment. Sparse event-based vision transformers provide energy-efficient event-driven computation, yet their continual learning capabilities remain largely unexplored. We here introduce sLoTh, a parameter-efficient continual learning framework for pretrained sparse event-based (spiking) vision transformers. sLoTh freezes the backbone and restricts plasticity to scalable-efficient low-rank attention updates (seLoRA) and shared neuronal threshold modulation, enabling adaptation without replay buffers by updating less than 1% of model parameters. Experiments across CIFAR-100, Tiny-ImageNet, ImageNet-100, and ImageNet-R with up to 100 tasks demonstrate competitive rehearsal-free performance in class-incremental learning and online continual learning, while enabling approximately 6.5x lower energy consumption than conventional dense vision transformers.

Layout understanding, or the interpretation of element organization, is essential for document analysis, user interface (UI) creation, and graphic design. While recent vision-language models (VLMs) excel at interpreting atomic layouts composed of independent elements, they struggle with compositional layouts that require reasoning over visually entangled elements within hierarchical multi-layer structures. In this paper, we introduce a new task, compositional layout understanding, and present CoDeLayout, a VQA dataset of ~20K real-world multi-layer layouts annotated with compositional element pairs and design intent. Through empirical analysis on CoDeLayout, we identify two key challenges for existing VLMs: semantic drift between textual metadata and visual content, and structural ambiguity in hierarchical inter-element relationships. To address these challenges, we propose MASON, a post-training paradigm that integrates multimodal alignment (MA) and structural perception (SP). MA enhances element interpretation by grounding metadata-defined elements to their visual counterparts, mitigating semantic drift, while SP models layer-aware inter-element spatial relationships to improve hierarchical understanding and reduce structural ambiguity. Experiments reveal substantial gaps in existing VLMs: even the strongest baseline, GPT-o3, achieves only 79.68% accuracy, whereas Qwen2.5-VL 7B with MASON reaches 91.66%. Notably, MASON surpasses full-data Direct Finetune using only 30% of the training data and scales better with additional data.

Diffusion-based Video Virtual Try-On (VVT) achieves high visual fidelity through bidirectional spatio-temporal modeling, but complete-clip dependence incurs prohibitive latency and computational overhead in practical continuous deployment. Naively enforcing causality disrupts pretrained bidirectional priors and substantially degrades synthesis quality. We introduce LiveVVT, a rolling streaming diffusion framework that preserves bounded bidirectional modeling within causal recurrent generation. Within a fixed-size window, LiveVVT jointly denoises multiple video chunks under bounded look-ahead, preserving local bidirectional interactions while emitting one clean chunk per iteration. Beyond the window, two complementary memories sustain long-term consistency: a bounded temporal memory propagates recent dynamics and occlusion context, whereas a persistent global appearance memory, constructed once from the target garment and a frontal try-on keyframe, anchors garment details and dressed appearance throughout the stream. We further introduce a progressive distillation framework integrating bidirectional VVT learning, teacher-trajectory regression for causal few-step adaptation, and Collaborative Matching Distillation, which couples teacher-distribution matching with rolling flow matching on real videos to align optimization with recurrent inference. Experiments on paired and unpaired long-sequence benchmarks demonstrate superior generation quality over similarly sized models, with $26\times$ lower latency and $11\times$ higher throughput, enabling high-fidelity real-time streaming VVT.

Recent advances in Multimodal Large Language Models (MLLMs) have extended Image Aesthetic Assessment (IAA) beyond scalar scores toward interpretable critique and guidance. Yet existing benchmarks mainly assess intrinsic visual quality or fixed domain criteria, leaving open whether an appealing image is appropriate for a specific purpose, audience, cultural setting, or domain convention. We introduce AesCanvas, a unified suite with two complementary components: CritiqueCanvas with 519,136 instruction-response pairs from 54,300 images supports long-form, multi-dimensional critique across photography, painting, and virtual imagery, whereas ContextCanvas with 301 expert-reviewed use scenarios evaluates contextual aesthetic suitability in realistic use scenarios. Under a unified protocol, we evaluate closed-source frontier, open-weight general, and aesthetic-specific MLLMs. Results reveal a clear separation between critique generation and context-sensitive judgment: reference-based lexical and semantic metrics only partially capture critique quality, while aesthetic specialists remain competitive on selected critique metrics yet substantially lag strong general-purpose MLLMs on ContextCanvas. Further analyses show that aesthetic specialization does not reliably transfer to contextual suitability and that model decisions may fail to track or ground themselves in decisive contextual visual cues. These findings establish culturally situated, evidence-grounded suitability as a distinct objective for aesthetic modeling.

Expert-level financial question answering requires both grounded verification to catch numeric hallucinations and audit trails for regulatory compliance, attributes that standard single-pass RAG systems lack. We take a step toward this goal with Self-Improving RAG, a framework that decomposes document QA into three specialized agents (Retrieval, Reasoning, and Judge) coordinated by an orchestrator with feedback-driven self-correction. When the Judge Agent scores an answer below a dynamic threshold, the system triggers retry with escalated strategies: broader retrieval, more careful prompting, and relaxed acceptance criteria. We evaluate on FinanceBench (SEC filing QA), where Self-Improving RAG achieves 86% oracle-guided accuracy (measuring agreement with gold answers) with a 36.4% Lazarus Rate, recovering nearly 4 in 10 initially incorrect answers through targeted retry. A key finding is that a fixed retrieval pipeline with judge-driven retry achieves strong results without dynamic routing, providing full interpretability. Every decision is logged with confidence scores, enabling the audit trails required for regulated financial applications.

We present an extension and comprehensive real-world validation of Co-Scientist, a Gemini-based multi-agent system designed to accelerate end-to-end scientific research across hypothesis generation, experimentation, and manuscript generation. Moving beyond in silico hypothesis generation, this specialized configuration transitions Co-Scientist into an execution-grounded research partner advancing closed-loop scientific workflows across materials science, biology, and computer science. In materials science, Co-Scientist interfaced with a semi-automated chemical vapor deposition reactor to design a safe precursor route for MXenes; experimental execution produced a lamellar 2D material sharing key structural similarities with the Ti3C2Tx MXene lattice, although further experiments are needed to confirm the atomic structure. Leveraging Gemini 3 Deep Think for rapid, lab-in-the-loop execution, it also tailored growth recipes to laboratory constraints in minutes, enabling single-attempt growth of monolayer MoS2, MoSe2, and WS2 semiconductors. In biology, Co-Scientist predicted emergent swarming phenotypes of engineered E. coli across inducer (IPTG) gradients from sparse imaging data, quantitatively matching unpublished wet-lab morphological measurements. In computer science, Co-Scientist autonomously discovered an inference-time scaling architecture that outperformed six frontier models on HealthBench (Hard and Professional) while reducing potential clinical harm under blinded physician evaluation. Finally, a double-blind study of end-to-end generated papers with 30 domain experts across 450 reviews demonstrates that Co-Scientist's reliability modules reduce hallucination and plagiarism while improving research safety. Together, these results demonstrate progress toward closed-loop multi-agent scientific AI systems capable of accelerating real-world scientific discovery.

Enterprise deployments of autonomous AI agents inherit a control model built for human users and long-lived services, and the fit fails in three specific ways: agent principals are ephemeral, appearing and vanishing faster than provisioning; their actions are selected by a model rather than programmed, so the set of things they may attempt is not known in advance; and the population is discovered rather than provisioned, because anyone who can call an API can create one. We argue that governing such agents is a runtime problem -- not a model-alignment problem and not a build-time problem -- and we derive five primitives from the questions that must be answered before an action takes effect and after it has: discovery, identity, governance, attestation, and supply chain. For each we state what fails if it is absent and why the others cannot structurally supply it. We describe an implementation in which an agent's action is mediated against policy before it takes effect, authorised against a per-tenant action vocabulary, and recorded in a hash-linked signed ledger a third party can verify with the vendor out of the loop. We report what the architecture costs: the enforcement point sits on the request's critical path, identity requires a sidecar per workload, and fail-closed mediation converts availability incidents into denial. We are explicit about implementation status: four primitives are built and running in private pilots, and the fifth is built as separate tooling and not yet integrated into the request path. We keep it in the set deliberately: a five-part decomposition that exactly matches what its authors happened to build is not a taxonomy but a description of a codebase.

Detecting AI-generated text (AIGT) remains challenging because existing approaches rely on token-level statistical signals or independent stylometric features, causing them to overfit to specific generators and fail under distribution shift. We identify a structural signal at the sentence-pair level: LLMs produce inter-sentence transition variance that deviates from human writing through inflated variance driven by recurring similarity bursts at paragraph boundaries and templated transitions. We formalize this as Relational Over-Regularization (ROR) and validate it across four benchmarks (p < 0.001). The central contribution is this relational problem formulation, not a novel GNN architecture; CSFG is one concrete instantiation for operationalizing ROR. To exploit this signal, we propose the Cross-Source Stylometric Fingerprint Graph (CSFG), a graph-based framework that encodes positional, sequential, semantic, and transition deviation signals as learnable GNN edge features. The per-edge signed deviation δ_ij operationalizes ROR without hand-crafted thresholds and acts as a false-positive calibrator. CSFG achieves 97.14% accuracy under binary detection, outperforming the strongest graph-based baseline by 11.14 pp, with a false-positive rate of 1.57% and robust generalization to unseen LLMs in the inflated-variance regime; detection degrades for generators whose transition variance falls at or below the human baseline.

Despite the growing number of public datasets, annotated medical images remain scarce. Supervised learning methods achieve strong performance on many benchmarks, however require large amounts of labeled data, which are costly and time-consuming to obtain in the medical domain. To address this limitation, contrastive self-supervised learning (SSL) has emerged as a promising alternative for learning useful representations from unlabeled data. In this work, we investigate two SSL frameworks, SimSiam and SimCLR, for retinal disease classification from fundus images. We focus on understanding how augmentation strategies and training parameters influence representation learning under resource-constrained settings. Given limited data and computational capacity, we explore the feasibility of training SSL models with small batch sizes incorporated with retinal-specific augmentation techniques. Through a series of experiments, we assess the quality of learned representations via linear evaluation and fine-tuning across downstream tasks, including multi-disease classification and diabetic retinopathy grading. Our results show that tailoring augmentation strategies to the characteristics of retinal images plays a critical role in improving performance. Even under constrained settings, lightweight SSL frameworks can learn transferable representations that reduce dependence on large annotated datasets and achieve competitive results.

Recent large language models achieve strong performance on complex reasoning tasks, where reinforcement learning with Group Relative Policy Optimization (GRPO) has emerged as a leading paradigm for optimizing models on self-generated trajectories. However, the on-policy nature of GRPO bounds the model to the reasoning skills it can already produce, restricting to learn more advanced capabilities. Prior works inject privileged reasoning traces from a stronger teacher policy to guide training, yet these traces are inherently out of distribution with respect to the student policy. We observe that this mismatch between on-policy and off-policy causes gradient clipping on semantically critical reasoning tokens, ultimately rewarding correct answers while leaving the reasoning that justifies them unlearned. Hence, we propose \textbf{Echo-GRPO}, a framework that lets the model reason in the words it speaks. Rather than imitating low-probability privileged traces from the teacher model, Echo-GRPO rewrites them into the student policy's own \textit{idiolect}, that is, its own characteristic vocabulary and expression patterns, while preserving their semantics via Dual-Reference Decoding. We instantiate this framework as \textbf{VideoEcho-R1} for video reasoning distillation, achieving consistent improvements across three multimodal LLM backbones and five benchmarks. Finally, we show that our idiolectal paraphrasing is a plug-in module that consistently improves both RL and supervised fine-tuning frameworks for reasoning distillation, demonstrating that policy-aligned supervision extends beyond GRPO.

Cooperative multi-agent reinforcement learning (MARL) faces significant challenges in maintaining robust coordination under noisy observations. Although observation disturbances are often introduced independently across agents, their downstream effects on cooperative decision-making can become structured through underlying cooperation structures. We characterize this phenomenon as structured noise effects, where noise-induced decision effects exhibit local correlation among agents with stronger task-related dependencies while remaining globally heterogeneous across different agents and local structures. Existing robust MARL methods, however, rarely explicitly characterize or exploit such structure-dependent noise effects. To address this limitation, we propose SIGMA, a hierarchical collaboration framework that exploits cooperation structures to learn robust representations under noisy observations. SIGMA first organizes agents into adaptive local structures through density-based grouping and performs intra-group consensus aggregation to preserve shared task-relevant information while smoothing agent-specific representation deviations. Inter-group attention then adaptively integrates information across different groups to preserve global coordination while accommodating their heterogeneous contributions. Experiments on noisy-observation tasks in StarCraft II empirically validate the structured noise effects and demonstrate that SIGMA consistently improves robustness under observation noise while maintaining competitive performance in noise-free environments.

Large pretrained vision-language-action models dominate modern robot-manipulation benchmarks, but it remains unclear how much model scale is necessary for strong language-conditioned control, or whether fundamentally different control architectures can remain competitive at much smaller parameter budgets. We present PredVLA, a language-conditioned predictive-coding policy with only 0.68 million trainable network parameters and no robot-data pretraining, whose hierarchical generative recurrent dynamics predict visual features and proprioception while observations influence latent state only through online inference from the resulting sensory prediction errors. On LIBERO, PredVLA achieves an 86.9% mean success rate across the three short-horizon suites and 75.4% when the long-horizon suite is included. Under a controlled comparison using the same frozen front end, demonstrations, action decoder, and evaluation protocol, PredVLA achieves 3.7x and 7.4x mean success rates of parameter-matched Transformer and LSTM policies, respectively. The predictive-coding formulation also makes the contribution of observation-driven correction directly measurable: because observations influence the recurrent state only through prediction-error-based latent inference, disabling this inference yields an exact open-loop control condition. Together, these results show that a sub-million-parameter recurrent generative policy can achieve strong performance on modern language-conditioned manipulation benchmarks while providing an explicit mechanism for prediction-error-driven online state correction.

Long autoregressive video generation faces a fundamental memory challenge: with a finite attention window, a model must decide which information from an ever-expanding history to retain. Existing methods organize memory temporally, preserving recent frames while compressing or discarding older ones. We instead propose RECAP-Forcing, organizing memory by appearance novelty. A long video is not merely a sequence of frames, but an evolving cast of subjects, objects, and scenes whose identities must remain consistent over time. We organize memory by retaining the KV cache associated with newly appearing content--such as entering subjects, disoccluded regions, and newly introduced scenes--at the moment it first becomes visible, prioritizing novelty over recency. Memory should scale with the amount of newly introduced content, rather than with video length. This appearance-indexed memory makes long-range consistency an explicit property of the memory structure. Our framework unifies two mechanisms under this single principle. At the beginning of a video, when all visible content is novel, an attention sink preserves the initial scene. As the video evolves, an optical-flow-based novelty bank extends the same principle by selectively retaining newly revealed content. As a training-free inference method with no additional learnable parameters, RECAP-Forcing consistently improves visual quality and semantic fidelity across multiple strong baselines and outperforms existing memory methods.

Maritime vessel detectors often face scenes where hulls are small, low-contrast, or blurred, while wakes are longer and easier to detect. This creates a wake-reliance problem: detectors may miss slow or stationary vessels with weak wakes, or produce false positives on wake-like water clutter. We propose HullWake, a hull-first wake-second framework for robust maritime vessel detection. HullWake separates proposal-centered hull evidence from directional wake context, extracts wake cues with bidirectional proposal-anchored corridors, and suppresses wake-dominant predictions through wake response supervision, wake-attenuated consistency, wake-only confidence suppression, and hull--wake decorrelation. We also introduce a wake-oriented evaluation protocol covering weak/no-wake vessels, wake-like hard negatives, worst-group AP, and confidence drop after wake attenuation. Experiments are conducted on Curated-Wake, a wake-oriented maritime dataset of about 10,000 images curated from Ships/Vessels in Aerial Images, the SMD benchmark, and SeaDronesSee, with newly added detection- and segmentation-level wake annotations. Compared with box-only detectors and mask-supervised segmentation baselines, HullWake improves overall AP, weak/no-wake robustness, wake-like false positives, worst-group AP, and confidence stability after wake attenuation.

Many synthetic-image detectors produce accurate predictions but offer limited insight into how those decisions are formed. This paper introduces Hierarchical Channel Stacking (HCS), a compact framework for AI-generated image detection that converts intermediate CNN activations into a structured 60-dimensional representation organized across three progressively deeper backbone stages. HCS uses per-channel Level-1 classifiers and a Level-2 aggregator to produce image-level predictions while preserving explicit hierarchical structure for analysis. On a benchmark spanning GAN and diffusion generators, HCS achieves 86.7% accuracy and 86.7% macro-F1 on the held-out test set. Stage ablation shows that the full three-stage system outperforms reduced single-stage and two-stage variants, indicating that the hierarchy carries complementary predictive information. Stage-level contribution analysis further shows that, in the analyzed detector setting, fake GAN and fake diffusion images exhibit distinct stage-level contribution profiles. These results position HCS not simply as a compact detector, but as a structured framework for studying how synthetic-image detectors assemble evidence across representation levels.

Vision-Language-Action Models~(VLAs) have demonstrated significant promise in generalizing to complex, long-horizon robotic manipulation tasks. However, their performance remains brittle, as they are typically trained on trajectory-monotonic, failure-free demonstrations. This reliance on ``perfect" data leaves them unable to recover from common execution errors, such as a missed grasp, a dropped object, or an unexpected collision. In this paper, we propose FLARE, a novel framework that endows VLAs with robust error recovery capabilities through a ``Retry" and ``Reset" paradigm. First, we introduce a ``Retry" mechanism by injecting perturbation and bridging segments that decouple robot pose from environment state into demonstrations, enabling the policy to autonomously handle execution deviations. Second, to address critical, state-breaking (OOD) failures, we introduce a ``Reset" pipeline. We leverage an MLLM for offline failure analysis to automatically identify OOD states from execution videos. This analysis enables the efficient, targeted collection of a small library of object-centric ``Reset" skills, which are trained to restore the environment to a task-valid state. Our full framework integrates these learned policies. At inference, an online MLLM monitor arbitrates between task execution and ``Reset" skills. Experiments on challenging, contact-rich manipulation tasks show our approach significantly improves task success and robustness.

Event cameras capture pixel-level intensity changes with microsecond resolution to produce highly sparse asynchronous data streams. For visual perception in latency-critical environments, we propose a lightweight, training-free framework for discovery of moving objects based on spatio-temporal clustering. This framework is driven by two core contributions. First, a linear-time Spatio-temporal Probabilistic Event Filter (SPEF) that introduces an adaptive event acceptance threshold to distinguish salient motion structures from background noise. Second, an Event Morton Code Clustering (EMCC) module that bypasses expensive distance matrix computation to efficiently group events for unsupervised discovery of moving objects. On the E-MLB dataset benchmark, SPEF achieves the best denoising performance among classical filtering methods and remains competitive with learning-based approaches without requiring any offline training. On object discovery, EMCC achieves the highest overall accuracy and lowest execution time across the FRED and eTraM datasets, outperforming established density-based clustering baselines by a substantial margin. Overall, this work establishes a new performance benchmark for classical object discovery in event data, providing a highly scalable, training-free solution for resource-constrained visual perception. The code is available at https://github.com/PrathamShenwai/SPEF_EMCC

Decoding quality in diffusion multimodal language models (dMLLMs) depends heavily on the order in which masked tokens are committed. Existing confidence-based strategies prioritize locally easy tokens, but confidence does not necessarily reflect contextual usefulness. As a result, structurally easy tokens such as punctuation may be committed before informative semantic anchors, weakening context propagation and increasing error accumulation. We propose Information-Guided Frontier Decoding (IGFD), a training-free decoding strategy that ranks candidates using token confidence, neighborhood uncertainty, and structural commitment risk. IGFD encourages early commitment of reliable semantic anchors while delaying fragile structural tokens, improving contextual support during decoding. A dynamic candidate frontier further constrains token selection to locally expandable regions under the same decoding budget. The method requires no additional training, auxiliary models, or extra forward passes. Experiments across multimodal understanding, reasoning, grounding, and hallucination benchmarks show that IGFD consistently outperforms existing decoding strategies across the majority of benchmarks and diffusion MLLM backbones under identical decoding budgets.

Gait recognition has achieved remarkable progress, yet existing methods remain confined to rigid visual matching and often overlook the potential of natural language instructions for interactive retrieval. In this paper, we introduce Composed Gait Retrieval (CoGR), a novel task that retrieves a target gait sequence based on a reference sequence and a natural language modification query. To address the absence of existing datasets for this task, we design an automated annotation pipeline powered by large vision-language models (VLMs) to construct the first gait-language datasets: Language-Augmented CCPG and Language-Augmented CASIA-B. Building on this, we propose ComposeGait, an identity-anchored composition framework designed to prevent the identity drift that arises when generic composed retrieval follows the instruction but returns the wrong person. Its Part-aware Identity Adapter (PIA) aggregates multi-frame, part-aware identity evidence into a sample-specific ID token. We inject the ID tokens into both branches of a shared Q-Former to preserve identity, while excluding the ID-token outputs from the final retrieval embeddings. Joint identity and task-adapted composed-retrieval objectives optimize this space end to end. We evaluate ComposeGait on both benchmarks and show that it achieves the best R@1 among the compared methods, reaching 72.38% on Language-Augmented CCPG and 83.61% on Language-Augmented CASIA-B. These results establish ComposeGait as a strong baseline for CoGR. The datasets and code will be made publicly available.

This report presents a framework to help organisations, policymakers and researchers reason about the risks that emerge when AI agents interact with each other, how those risks change as interactions cross organisational boundaries, and the controls that may help address them. As organisations deploy AI agents, those agents will increasingly interact with each other: inside the organisation, with the agents of partners, customers and suppliers, and with unknown counterparties on the open internet. Failures can emerge from the interactions themselves, and once those interactions cross an organisation's perimeter, no single organisation can fully see, control or govern them. The report introduces three deployment tiers, defined by the minimum common governance binding any two interacting agents: singular governance, where one organisation governs every agent; federated governance, where multiple organisations deploy into a shared environment under agreed rules; and open environments, where agents operate with no central authority and shared standards are adopted voluntarily if at all. Within each tier, the report examines risk factors, failure modes and available controls. It identifies who is positioned to apply the controls, and where no actor is positioned to act, it characterises the gap and the collective action required to close it.

Open-vocabulary semantic segmentation (OVSS) aims to segment image regions corresponding to arbitrary text queries. Although the Segment Anything Model (SAM) is a powerful foundation model for segmentation, its standalone performance on OVSS remains limited. Existing methods therefore often use SAM to refine coarse masks predicted by other models, but this strategy is unreliable when the initial masks are inaccurate. In this work, we argue that more reliable segmentation can be achieved by exploiting SAM as a region expansion module guided by accurate object points (i.e., seeds) rather than inaccurate coarse masks. Inspired by classical seeded segmentation, we reformulate OVSS as text-guided seed localization followed by seed-based region expansion. To realize this idea, we propose Text-to-Seed (T2S), a training-free framework that leverages the text-to-region correspondence of Stable Diffusion to generate attention-based seed points for target categories described by text. These sparse seeds are then used as point prompts for SAM to produce full object masks. Without task-specific training or additional annotations, T2S achieves strong performance on standard OVSS benchmarks, demonstrating the effectiveness of combining semantic grounding with seed-driven spatial segmentation.

Achieving stable, sustained grasping with soft robotic hands remains a fundamental challenge. Compliance enables safe and adaptive contact, yet the intrinsic viscoelasticity of soft polymers leads to stress relaxation and a continuous decay of grasping force during holding. Inspired by human grasping, which combines phase-dependent stiffness regulation with continuous sensing and feedback, this paper presents an integrated structure--perception--learning framework. We develop a variable-stiffness soft gripper that uses onboard vision and infrared thermography to track deformation and the temperature field in real time, preserving continuous tracking of the interaction state. To mitigate relaxation-induced force decay, we propose a temperature-coupled viscoelastic force representation, together with a physics-informed learning model, to reconstruct the force trend and provide explicit compensation during holding. Experiments show that, in a 280s force-controlled grasp-and-hold task, the proposed method maintains the desired force with a mean absolute error of 0.066N, outperforming fixed-aperture and instantaneous-only baselines by 80% and 95%, respectively. Overall, the results support a mechanism--AI co-design view: mechanisms shape feasible interactions, while learning compensates remaining uncertainty in viscoelastic dynamics, together enabling stable, sustained grasping.

Model selection becomes particularly challenging under strong predictor dependence and model-class uncertainty, especially when there are exponentially many models. We propose a Descriptive-Complexity Information Criterion (DCIC) that regularizes large candidate model collections through Kraft-admissible code lengths. Under sub-Weibull noise, we establish selection consistency through approximation-error separation without relying on RIP-type conditions, together with nonasymptotic oracle risk bounds that remain valid under model misspecification. The same coding principle places heterogeneous classes on a common complexity scale at a small additional class-identification cost. This extension yields class--model recovery under suitable identifiability conditions and risk adaptation across classes. We further develop a complexity-guided search path that makes the computation--statistics trade-off explicit. Large penalties yield polynomial-size retained search regions with high probability, whereas smaller penalties sharpen the oracle risk benchmark. Numerical experiments illustrate stable support recovery and favorable estimation performance under strong dependence and model-class uncertainty.

This study provides a comprehensive benchmarking of conventional machine learning (ML), ensemble learning, deep neural networks, recurrent architectures, Transformers, graph based models, and hybrid ensemble deep learning approaches under complementary renewable energy scenarios. Three datasets are considered: a large scale WEC dataset, a 16 WEC dataset, and operational 10 min SCADA measurements at the Penmanshiel wind farm. For structured WEC layout data, tree ensembles exhibited a clear advantage over conventional ML and neural predictors because randomized partitioning and boosting efficiently captured nonlinear layout power interactions without requiring explicit feature representation learning. The Extra Trees was the strongest model, achieving considerable results. Relative to the MLP baseline, this corresponds to an approximately 63.7% reduction in MAE, demonstrating the suitability of randomized tree ensembles for high dimensional structured WEC data. Also, STGCN reduced the MAE to approximately 167.0 kW and achieved R = 0.93 by explicitly learning spatial and temporal turbine interactions. The best overall forecasting accuracy was obtained by the RF BiLSTM hybrid, with an MAE=150.5 kW. Compared with standalone LSTM, this represents an approximately 75% reduction in MAE, while improving on STGCN by approximately 10.0%. Finally, the experiments reveal that no single AI architecture is universally optimal: randomized and boosted ensembles are particularly effective for structured WEC surrogate modeling, graph networks become advantageous when explicit spatial interactions dominate, and ensemble recurrent hybrids provide the strongest balance when nonlinear tabular relationships and temporal dynamics coexist.

In university academic advising, identical questions can require different answers depending on a student's department, admission cohort, and degree program, causing profile-blind retrievers to surface plausible but inapplicable evidence. We present proFILL, a method for transforming hoBIT, our college's current rule-based advising chatbot, into a profile-aware retrieval-augmented generation (RAG) system. Rather than requiring a complete user profile upfront, proFILL progressively acquires only the profile attributes needed for each query, guided by both the query intent and the initially retrieved evidence, and uses them to condition retrieval over a profile-aware index. Extensive experiments and a human preference study show that proFILL outperforms diverse RAG baselines, is preferred by target users, and remains effective with open-weight models for cost-effective on-premise deployment.

Following clinical decision pathways (CDPs) defined by clinical practice guidelines is essential for safe and reliable medical decision-making. However, existing medical large language model (LLM) benchmarks mainly evaluate final-answer accuracy, providing limited evaluation of models' ability to adhere to guidelines. To address this gap, we introduce MEGA-CDP, a benchmark for evaluating whether medical LLMs can generate guideline-adherent CDPs using provided guidelines as references. MEGA-CDP is constructed from 2,274 English and Chinese clinical practice guidelines through a guideline-to-case pipeline, yielding 42,353 clinical cases with explicit reference CDPs. It supports both single-turn vignette and multi-turn interactive settings, and introduces a CDP-oriented evaluation framework for measuring pathway consistency. Experiments on 16 representative LLMs show that reliable clinical decision support remains challenging for current models, demonstrating the need for CDP-oriented evaluation and the value of MEGA-CDP for advancing guideline adherence in medical LLMs.

Biomedical knowledge graphs (KGs) offer structured medical knowledge that can ground large language model (LLM) reasoning in clinical diagnosis application, yet how KG signal should be integrated into LLMs remains an open question. We present a systematic study spanning five KG task formulations, three training paradigms, two KGs, and three base LLMs. At the task level, all paradigms improve over the non-finetuned baseline, but methods with comparable in-domain accuracy show substantially different knowledge transfer behavior. We introduce Gradient Intervention Density (GID) and Gradient Distortion (GD) to measure how broadly an optimizer modifies the pretrained model. GID and GD together reveal a clear divide: KG-judgment training under KL regularization produces sparse, localized updates (a regime we term as surgical alignment), while task-specific SFT produces dense ones. A controlled ablation shows that the objective and KL contribute to sparsity independently, and the paradigms that produce sparse updates also improve reasoning quality, even when their in-domain accuracy is lower than task-specific SFT. Assessing KG-LLM integration thus requires complementing accuracy with optimization-geometry diagnostics. Our implementation can be found at https://github.com/LARK-NLP-Lab/Surgical-Alignment.

Humans traverse complex terrain over long distances without losing balance, whereas perceptive humanoid policies become fragile as perception and control errors accumulate. We present SOLO, a unified framework addressing two compounding causes of this long-horizon fragility: dense terrain reconstruction smooths action-critical details, and pointwise imitation lacks temporal credit assignment. Its Query Reconstructor (QR) uses Fourier-encoded cell queries to retrieve spatially specific evidence from depth-proprioception tokens, preserving sharp terrain boundaries. Trajectory-Aware MSE (TA-MSE) Distillation adds next-state teacher-student disagreement to the PPO reward, enabling Generalized Advantage Estimation to propagate future disagreement penalties to preceding actions. In simulation, QR reduces height-map L1 error by factors of 3.3-4.0, while TA-MSE surpasses PPO and MSE+PPO in curriculum progression. On stress-test terrains, SOLO achieves 97.5% mean traversal success and 96% stepping-stone success, versus 75.0-75.6% and 0-3% for dense-reconstructor variants. Deployed zero-shot with only a chest-mounted depth camera and proprioception, SOLO completes a continuous 1.5-km outdoor route and an indoor mixed-terrain course. Project page: https://sunpihai-up.github.io/solo/

Self-evolving language models have recently emerged as a promising path toward superintelligence, with the advantage of reducing the cost of human supervision. While considerable progress has been made in verifiable domains, self-evolution in unverifiable domains remains substantially less explored. We propose Judge co-adaptation from Zero data (J-Zero), a unified Challenger--Solver--Judge co-evolution framework that supports self-improvement across both domains. The Challenger and Solver co-evolve through an adversarial interaction: the Challenger generates increasingly difficult tasks, while the Solver learns to produce higher-quality responses to them. In parallel, the Judge co-adapts using preference pairs whose ordering is known in advance from how each response was produced, i.e., the Solver's answer over the Challenger's, and its decomposed-and-recombined answer over its one-shot answer, rather than from the Judge's own scores. J-Zero outperforms the baselines by an average of 4.2 points on verifiable and 8.0 points on unverifiable domains, and continues to improve through at least ten iterations, whereas the baselines degrade after two.

Low-bit quantization offers a promising avenue for reducing the computational and memory demands of Multimodal Large Language Models (MLLMs). Recent hardware support for low-precision formats, ranging from MXFP8 to ultra-low-bit formats such as MXFP4 and HiF4, has accelerated research into efficient MLLM training and deployment. In this work, we present a systematic study of these quantization schemes in representative MLLMs that span both video generation and reasoning tasks. Our analysis shows that MXFP8 achieves near-lossless performance, whereas aggressive 4-bit quantization leads to significant degradation. Through extensive ablations, we identify activation quantization as the primary source of this performance loss, contributing substantially more than weight quantization. Motivated by this observation, we propose Residual Fallback Quantization (RFQ), a lightweight activation reconstruction framework that supplements the primary ulta-low-bit activation representation with an auxiliary quantized residual pathway. By explicitly modeling and compensating for quantization errors, RFQ improves activation fidelity while preserving the efficiency advantages of ultra-low-bit computation. RFQ requires no architectural modifications and incurs negligible computational overhead. Extensive experiments on Wan2.2 and Qwen3-VL demonstrate that RFQ consistently recovers a substantial portion of the performance lost under the quantization of MXFP4 and HiF4, significantly narrowing the gap to BF16 baselines across both generation and 4 reasoning benchmarks. Our findings establish activation quantization as the dominant bottleneck in ultra-low-bit MLLMs and highlight residual-based activation reconstruction as an effective and practical strategy for robust 4-bit deployment.

Diffusion multimodal large language models (dMLLMs) have recently emerged as a new decoding paradigm for multimodal generation. Starting from a fully masked sequence, dMLLMs progressively decode the sequence by unmasking a subset of the remaining masked positions at each step. Since the selected tokens serve as the prediction context for subsequent steps, deciding which tokens to decode is crucial to the quality of the final output. The most common strategy prioritizes tokens based on a certainty measure that tends to favor tokens frequently observed in the training data. Recent approaches instead order tokens according to their influence on subsequent predictions, but do not explicitly account for the input image. We propose the Visual Information-Guided Sampler (VIG-Sampler), which prioritizes tokens based on their attention to image tokens. We further impose a constraint that penalizes candidate tokens whose image-attention distributions are similar to those of previously selected tokens, thereby increasing the information gain of the decoded subset. Extensive experiments on 7 captioning and VQA benchmarks with 3 open-source dMLLMs demonstrate the effectiveness of VIG-Sampler, which outperforms the Info-Gain Sampler by an average of 19.3 CIDEr points across the captioning benchmarks and surpasses it on COCO Caption while using only half as many decoding steps.

This work introduces Configured Failure Trapping, a novel backdoor attack task against Vision-Language-Action (VLA) models, which aims to activate attacks through stealthy textual triggers and induce configured failure modes. Unlike prior backdoor attacks that treat any task failure as a successful attack, Configured Failure Trapping requires the attacker to control how the robot fails (e.g., causing the robot to grasp with a specified positional offset), making it substantially more challenging and hard to detect. To support the new task, we propose an effective data engine for synthesizing high-quality target trajectories and an automated suite for measuring configured-failure fidelity. Then, based on this foundation, we construct two new benchmarks, namely Trap-LIBERO and Trap-RoboTwin, that instantiate Configured Failure Trapping across four representative failure modes. To address this task, we identify sparse action deviation as a critical challenge and accordingly propose a novel method named TrapVLA, which explicitly learns trigger-induced action residuals to steer the policy toward the configured failure behavior. Extensive experiments across simulation benchmarks and real-world robotic settings show that TrapVLA effectively injects configured failure modes into VLA models while largely preserving performance on clean data. Project page: https://john-liua.github.io/TrapVLA/

Diffusion large language models (dLLMs) offer a promising alternative to autoregressive generation by decoding multiple tokens in parallel through iterative denoising. However, increasing decoding parallelism often degrades generation quality, as early errors can contaminate later contexts. Revocable decoding mitigates this issue by re-evaluating decoded tokens and remasking unreliable ones, but existing methods overlook that unreliable tokens may also corrupt the verification context itself. We identify this failure mode and propose Dependency-Aware Revocable Decoding (DARD), a training-free framework that separates tokens into masked, candidate, and unmasked states. DARD verifies candidate tokens using a selective context that excludes less reliable tokens and adaptively regulates their influence on subsequent decoding. Experiments across 12 textual and multimodal benchmarks on 3 open-source dLLMs show that DARD consistently improves the speed-quality Pareto frontier over recent revocable decoding methods, achieving a 2.71$\times$ speedup and a 4.35-point CIDEr score gain over Saber on Flickr30K.

Agentic (tool-using) language models are mainly trained on tool-call traces and agent trajectories during post-training. These data provide direct behavioral supervision, but producing them requires task environments, execution, and verification, making broad tool and task coverage expensive. Publicly available skills offer another source of training data: they encode reusable tool semantics and workflows but are typically used only as inference-time context. We introduce Skill Pre-Training (SPT), a mid-training method that applies causal language modeling to SkillCorpus, a collection of public multi-file skill packages, optionally mixed with general data. To preserve relations among files within each package, we also introduce Reference Insert, a reference-aware assembly strategy that places supporting files near their mentions in the primary instruction. Experiments across multiple model scales and post-training recipes show that SPT consistently improves agentic performance over mid-training on general or trajectory data, while largely preserving general performance. Data mixture experiments show additional benefits from combining skill data with general annealing corpora. These results indicate that skill packages are a valuable data source for pre-training agentic language models.

Autonomous agents are increasingly adopted to complete complex, multi-tool workflows in real-world settings. However, existing benchmarks typically separate tasks by application or capability and evaluate agents in environments that are cleaner and more stable than those encountered in practice. We introduce DuMateBench, a real-session benchmark reconstructed from anonymized and privacy-screened user sessions collected from a large-scale production agent platform. Each task preserves the relevant pre-solution interaction history, persistent configurations, and workspace state, and is then validated through human verification. The resulting benchmark comprises 200 tasks spanning 8 broad scenarios and 17 fine-grained capability categories, with most tasks requiring multiple capability coordination. We execute these tasks in isolated Docker containers injected with three forms of real-world environmental complexity: Insufficient, Unstable, and Noisy, and assess performance using a hybrid deterministic and LLM-as-Judge evaluation protocol. Experiments across five representative autonomous-agent frameworks paired with four state-of-the-art LLMs reveal substantial gaps in strict task completion. Complementary robustness, efficiency, and diagnostic analyses further show that performance under environmental perturbations is jointly shaped by the capabilities of the LLM and the surrounding agent framework. The code and data are publicly available at https://dumatebench.com/.

Robot policies deployed in the wild should have the capability to continually learn new tasks without forgetting existing behaviors. A common approach to combat such catastrophic forgetting is to train on new task data with a replay buffer of previously learned task data. Although this buffer is commonly sampled randomly from all prior experiences, we show that a small set of these experiences contributes greatly in anchoring past performance. We call these experiences Memory Anchors. We identify Memory Anchors in regions where representations of new-task observations collapse onto those of old-task observations even though the tasks require conflicting actions, like when a familiar object must be manipulated in a new way. Rehearsing old data in this region plays a key role in preventing destructive overwriting of past task knowledge, serving as this critical Memory Anchor role. Excluding only 10% Memory Anchors before sampling the buffer leads to more than a 4.5x increase in catastrophic forgetting on the LIBERO benchmark suites. Conversely, enriching the replay buffer with Memory Anchors can decrease high-conflict task forgetting by 63% and enables successful continual learning of two task sequences on a real robot. Videos and additional visualizations can be found at https://robot-adaptation.github.io/MemoryAnchors

We present Chart2SVG, a multimodal large language model that converts static raster charts into structurally organized, semantically enriched SVGs that support programmatic editing. By incorporating chart-specific semantic tokens into a vision-language model, Chart2SVG captures both geometric primitives and their functional roles. To support robust structural recovery, we introduce Beagle+, a dataset of 33K canonicalized and structurally distilled chart samples. Our approach combines specialized training objectives with a rendering-aware post-training phase, producing SVGs that are both visually accurate and structurally consistent. To facilitate higher-level manipulations, we construct a Chart Structure Graph (CSG) that exposes visual dependencies, enabling tasks such as interactive exploration, chart repurposing, and layout reuse. Experiments show that Chart2SVG substantially outperforms baselines in reconstruction fidelity and downstream editing utility, advancing the development of intelligent and interactive visualization tools.

The advent of vision foundation models, notably the Segment Anything Model (SAM), has catalyzed significant advancements in natural image segmentation. However, their direct transfer to medical imaging remains severely bottlenecked by profound domain gaps, such as cross-modality and cross-center shifts. Existing Parameter-Efficient Fine-Tuning (PEFT) methods facilitate the adaptation of SAM to medical domains; nevertheless, they frequently suffer from performance degradation under severe distribution shifts. This vulnerability primarily stems from the implicit entanglement of heterogeneous frequency components within a shared low-rank subspace, which directly exacerbates sub-optimal structural alignment and localized boundary blurring. To overcome this representational bottleneck, we propose the Fourier-Adaptive Nonlinear Low-Rank Adaptor (FAN-LoRA), a novel frequency-decoupled fine-tuning architecture. FAN-LoRA explicitly separates the optimization space by employing a B-spline-driven low-pass branch for global structural alignment, synergistically coupled with a discrete Fourier high-pass branch for local textural compensation. Extensive experiments across three challenging cross-modality and cross-center benchmarks demonstrate that FAN-LoRA consistently outperforms state-of-the-art PEFT baselines. Compared to the strongest competitors, our method achieves consistent improvements in average Dice scores and notable reductions in boundary errors, while maintaining a compact module size without compromising computational efficiency.

In this paper, we explore multi-expert Conformal Risk Control (CRC) algorithms for pairwise LLM-as-a-Judge evaluation in open-ended dialogue. Our core insight is that multi-expert aggregation offers a complementary remedy to CRC: whereas CRC controls risk at the decision threshold through abstention, aggregation sanitizes the scoring function at its source. Guided by this, we first design two multi-expert CRC methods: Score Averaging and Decision Voting, which aggregate at the score and decision levels, respectively. While both strategies outperform single-expert methods on homogeneous expert panels, on heterogeneous LLM judges they remain risk-valid but recover only limited coverage, because a uniform threshold cannot match the experts' distinct scoring scales. To resolve this issue, we further propose Marginal-Calibrated Conformal Consensus (MC3): it captures distinct per-expert scales via initial threshold ratios, while jointly tuning a unified decision function $C_t(x)$ applied identically in both calibration and test, thereby preserving exchangeability. To evaluate our framework, we construct Panel, a 1,800-pair human pairwise-preference benchmark for open-ended dialogue. It is built on responses generated by four open-weight LLMs over dialogue contexts from three domains (ESConv, MSC, DREAM), with full logit access. In experiments, we find that both Score Averaging and Decision Voting substantially improve accuracy and acceptance rate on homogeneous panels. Notably, MC3 extends these gains to heterogeneous panels by accommodating distinct per-expert scoring scales across all three datasets.

We study energy-constrained adversarial patrolling on a graph, in which a battery-limited UAV defends a cluster of high-value targets against a strategic attacker who chooses when and where to strike. Unlike prior adversarial patrolling, the patroller must periodically return to a base to recharge; unlike prior energy-aware patrolling, it faces a self-interested adversary. Our central observation is that the remaining energy is a hidden state: the attacker never observes the battery directly, but observes the patroller's trajectory and can infer when a recharge excursion, and thus a vulnerability window, is imminent. We formalize the interaction as a zero-sum partially observable stochastic game and report a negative result on the solver side: neither independent deep Q-learning nor Neural Fictitious Self-Play reaches a stable equilibrium at this scale; each improves transiently and then collapses, with the co-trained thwart rate falling from about 0.25 to about 0.09 over training. Using a structural analysis independent of the learning dynamics, we show that achievable security rises monotonically with the energy budget, from zero below a threshold to about 0.7 when the budget is ample, establishing the energy budget as the primary determinant of defensibility. We set out the program the model is built to answer: whether an inference-capable attacker concentrates its successful strikes in the recharge window, and whether the defender can learn deceptive recharge timing to keep that window closed.

Visual intelligence seeks to perceive, interpret, and synthesize the visual world and is central to modern computer vision. Human-centered visual intelligence is especially demanding because it studies people as expressive, socially situated subjects whose meaning is rarely conveyed by appearance alone. It couples vision with audio and language across four representative tasks: human emotion recognition, human video generation, human voice cloning, and human video matting. Yet existing resources remain task-specific, providing modalities and annotations for individual problems rather than a shared foundation coordinating understanding and generation. This limits multimodal signal use and broader research. We address this gap with HUG-VIS, a unified benchmark for Human-centered Understanding and Generation in Visual Intelligence. It contains 8,400 seated half-body videos of 30 professional actors, each performing the same 280 emotion-action-prompt assignments under a controlled Mandarin studio protocol, with synchronized video, audio, text, and alpha mattes. We evaluate diverse open- and closed-source models across the four tasks under a unified zero-shot protocol using automatic metrics, criterion-specific mean opinion scores, and multiple cross-task analyses. Results show that (i) linguistic content dominates current emotion recognition, while purely visual affect recognition is weakest; (ii) in video generation and voice cloning, automatic metrics and human judgment agree overall but differ in their top rankings, requiring joint reporting; (iii) boundary fidelity under motion is the main remaining obstacle for human matting; and (iv) task difficulty varies across emotions, models, and metrics, with notable cross-task correlations. The dataset and results are available at https://github.com/GML-MMGroup/HUG-VIS.

Large language models often exhibit sycophancy, revising their answers to align with users when users push back. Such answer flips, however, can arise from different causes. One possibility is that the model simply aligns with the user's feedback in order to satisfy them. Another is that the feedback genuinely contains useful evidence, prompting the model to update its answer in a rational way. We distinguish them as Unsupported-Yielding and Rational-Updating. Prior work focuses primarily on suppressing Unsupported-Yielding, while overlooking its effect on Rational-Updating. We address this gap with a two-turn evaluation framework that measures the two behaviors separately. Across representative training-time and inference-time interventions, we find that anti-sycophancy methods often encounter a trade-off in which reducing Unsupported-Yielding can sacrifice Rational-Updating, and vice versa, even when the two objectives are optimized jointly. Mechanistic analysis suggests that the two behaviors share an internal substrate: the MLP neurons and attention heads driving them overlap substantially, and their associated steering directions are positively aligned. We further conduct a preliminary orthogonalized steering exploration, which yields modest, backbone-dependent selectivity gains. Overall, our results suggest that anti-sycophancy should be treated not as a simple suppression problem, but as a selectivity problem, where effective interventions should preserve Rational-Updating while reducing Unsupported-Yielding.

The Poppy Humanoid is an open-source, low-cost robot suitable for research and education in artificial intelligence. However, we are unaware of any published methodology that achieves reliable, unassisted bipedal locomotion on the standard Poppy hardware. This paper contributes a functional closed-loop walking controller for Poppy, based on the linear-quadratic regulator (LQR) framework for trajectory tracking. Starting with data collected from open-loop playback of a nominal walking trajectory, our proposed method learns a quadratic cost function for an LQR controller that substantially improves the reliability of the motion. The closed-loop controller is validated empirically, demonstrating statistically significant improvements in walking performance compared to open-loop trajectory playback.

Navigation in unknown environments to find unforeseen objects has become increasingly feasible with capable vision and language foundation models. However, these models also introduce non-negligible inference latency, which becomes an important concern when agents must operate continuously in the real world. Most state-of-the-art methods are still developed in synchronous simulators, where the environment waits for the agent to act and inference time is effectively free. As a result, agents are often designed around the sequential execution of perception, reasoning, and action, with little regard for time constraints. Under real-time execution, where wall-clock time counts towards the task budget, the inefficiencies of these architectures become clear. We show that recent zero-shot object navigation methods suffer consistent performance degradation under such realistic timing conditions. Motivated by this observation, we propose RTNav, a simple but effective architecture that treats inference latency, asynchronous environment stepping, and bounded compute as explicit design considerations. Evaluated on real-time variants of HM3D-v1, HM3D-v2, and HM3D-OVON, RTNav improves the success rate by up to 11% and the Success weighted by Completion Time by up to 5.1 points over prior work.

Multi-agent large language model systems are widely reported to beat single-model baselines, but the evidence is mixed, and comparisons are usually confounded: pipelines change token budgets, tool calls, and prompts simultaneously, so an aggregate gain rarely reveals what actually helped. We investigate the effect of introducing the manager-worker scaffold over a shared filesystem workspace, with no training and no per-benchmark tuning, measured against the same model answering in a single pass. Across nine models -- five open-weight, spanning 9B to ~2.8T parameters, and four frontier closed models -- on the 100 latest hard LiveCodeBench problems, the scaffold's benefit is real but conditional: large and statistically significant for some (Qwen3.8-27B +23.4, GPT-5.6-Luna +10.6 and GPT-5.6-Terra +8.0, each over five paired passes; Kimi-K3 +30.4 and Minimax-M3 +11.0 over five paired passes with reasoning off, both at $p < 10^{-4}$, and +42 and +12 in a single pass at a 128k cap) and null or negative for others (Qwen3.6-35B -1 to -9 with reasoning off). With the manager, Opus-5 achieves the highest score in the study at 91% in one pass. Running a manager roughly triples the token bill, but it buys accuracy more cheaply than moving to a larger model does: GPT-5.6-Terra with a manager nearly matches Fable 5's single-call accuracy (85.0 against 87.4, $p = 0.59$) at a fifth of the price (\$11.71 against \$61.11 per 100-problem pass, $p < 10^{-4}$), and the Qwen-27B arm does it for \$51.75 on weights anyone can self-host. Our transcript analysis finds several mechanisms behind the gains, of which two recur: context management, in which short worker calls and shared notes organize state and reduce truncation, and problem decomposition. Improvements are modest for large models with reasoning enabled, but larger for some models with reasoning disabled and for smaller models with reasoning enabled.

Finetuning has become the gold standard for refining existing behaviors and inducing new ones in language models, yet it often remains unclear exactly which behaviors emerge during this process. As models grow ever more capable, understanding finetuning better becomes increasingly important, particularly since unwanted behaviors may arise during finetuning. In this paper, we introduce Diff Mining, a simple yet effective framework for identifying what a finetuned model has learned by comparing its logits to those of its base model. Diff Mining effectively surfaces salient tokens that are amplified in the finetuned model, serving as a fingerprint of its training -- even on text unrelated to the finetuning domain. Unlike many existing model diffing methods which require model internals, Diff Mining only needs access to output logits and scales to large models. The framework consists of two modular stages: (i) extracting per-context logit differences between the finetuned and base models on a reference corpus, and (ii) aggregating the resulting signals to construct an interpretable token set representing the finetune. For aggregation, we explore both a simple Top-K frequency method and a Non-negative Matrix Factorization (NMF)-based approach for disentangling multiple finetuning objectives into distinct token clusters. Empirically, Diff Mining succeeds across diverse settings: on finetune domain detection, it significantly outperforms state-of-the-art model diffing methods both in identifying relevant tokens and in downstream performance when an interpretability agent is given access to the extracted token set; on models with injected biases, it identifies more than one third of the biases without targeted probing. Overall, our framework shows promise in developing auditing tools to detect finetuning objectives.

Recent advances in Large Language Models (LLMs) have shown that increased inference-time reasoning can improve performance on complex tasks. However, many existing approaches rely on fixed or preallocated reasoning controls, such as fixed token budgets, pre-execution difficulty estimates, or activation-space interventions, and are often evaluated on standalone reasoning benchmarks rather than full agentic workflows. These assumptions may not hold in agentic AI systems, where reasoning requirements evolve dynamically through planning, tool use, memory retrieval, and agent-to-agent interactions. Consequently, reasoning can become either excessive or insufficient, resulting in unnecessary computation, increased latency, planning drift, excessive tool use, or incomplete solutions. We argue that a major challenge for next-generation agentic AI is not merely how much reasoning a language model should perform, but how it should allocate reasoning according to evolving task demands. We characterize over-reasoning and under-reasoning as recurring failure modes of misallocated reasoning and evaluate them on MATH-500 and the GAIA public validation benchmark. Using tool-decision latency, token consumption, token-limit exhaustion, and answer correctness, our results suggest that cases classified as over-reasoning are associated with higher computational cost without proportional accuracy gains, whereas cases classified as under-reasoning are consistently associated with incorrect or incomplete solutions. These findings motivate future research on adaptive reasoning mechanisms for agentic AI.

Code-switching is pervasive in bilingual African conversation, yet most ASR systems assume monolingual input and are evaluated on curated monolingual benchmarks. We present AfriSwitch, a 61.36-hour human-transcribed benchmark of in-the-wild code-switched speech spanning 16 African languages and language varieties, released with switch-level English span tags, perutterance Code-Mixing Index (CMI), and switch-point counts. Corpus statistics show that mixing behaviour varies widely across African languages along two largely independent axes: how often speakers alternate, and how balanced the mixture is. No single scalar captures how code-switched a language is. Benchmarking five open and commercial multilingual ASR systems zero-shot yields word error rates far above published monolingual figures for the same languages, with the best system averaging 35.93% WER and no system falling below 24% on any language. Africa-targeted training, not model scale or nominal language coverage, best predicts performance.

Voice agents must call tools and hold multi-turn dialogue entirely through speech, yet the dominant paradigm trains them in text. Existing frameworks either cascade TTS and ASR around a proprietary voice API, where gradients cannot flow and per-call cost makes on-policy reinforcement learning prohibitive, or stay in text: they measure voice agents but cannot improve them. We present SpeechGym, an audio-native agentic environment in which two omni-modal models converse in native audio, with no external ASR or TTS and no API boundary, over the unmodified tasks, tools and success check of an established text agentic benchmark, so that the interaction modality is the only variable and the loop stays local and trainable end to end. Audio agentic capability does not follow from audio understanding. The failures speech introduces are perceptual rather than reasoning deficits: the agent picks the right tool and the right argument slot but fills it with a value misheard from the waveform, and that single error cascades into a failed call, a retry of the same call, and a wasted step budget. A second failure is behavioural: under an insistent caller the agent performs an unauthorised write and ends the episode believing it helped. Both are trainable, because the environment labels them for free: a call with a misheard argument fails against the database while a correct one succeeds. The obstacle is sparsity, not signal. Outcome-only GRPO is gradient-starved here, since almost every rollout group fails identically, while a per-turn process reward crediting each successful tool call restores variance to nearly every group. Trained this way, the agent transfers with no further tuning to an independently implemented voice benchmark, more than doubling task success and carrying an open-weights model from last place to second on that leaderboard, while using fewer turns and tokens than before training.

Quantum systems are invariably coupled to a surrounding environment. A common theoretical method to simplify the problem is to trace over the environment under a Markov approximation, which assumes that the environment holds no memory over the timescales that the system evolves under. In interacting many-body systems, it is often not trivial to extract which of the multiple collective timescales are relevant. We consider a time-delayed waveguide QED setup with a single excitation. By fixing the maximum propagation time and increasing the number of emitters, we show that, even for small delay times, collective effects are sufficient to cause non-Markovianity. The impact of memory is intrinsically state-dependent. For superradiant states, the relevant system timescale is the superradiant lifetime while the relevant bath timescale is the end-to-end delay time. For subradiant states, both timescales depend on the structure of the specific state. Our results demonstrate the importance of prudently making Markov approximations in quantum many-body systems, and highlight potential pitfalls to avoid in scaling up quantum devices to large system sizes.

SVD-based low-rank compression has become a fast-growing direction for reducing the memory and computational cost of large language models (LLMs). However, meaningful comparison across existing studies remains difficult as prior evaluations use varied benchmarks, inconsistent ratios, and diverse setups, often failing to isolate low-rank effects from auxiliary techniques. As a result, it remains unclear whether reported gains reflect method-level improvements or differences in evaluation protocol. This lack of comparability highlights the need for a unified, reproducible evaluation platform. To address this problem, we present LowRankArena, a standardized evaluation platform for SVD-based LLM compression. LowRankArena unifies task versions, uniform-precision compression budgets, comparison regimes, and inference measurements, and provides a reproducible pipeline with over 3 TiB released compressed checkpoints. Using LowRankArena, our aligned audit of five representative SVD methods reveals that prior findings are highly conditional under standardized protocols: clear leaders and performance tiers shift across backbones and keep ratios, multiple-choice accuracy can hide large perplexity degradation, and nominal low-rank savings yield workload-dependent and often limited end-to-end speedups. Our code is available at: https://github.com/Zishan-Shao/lowrankarena.git.

Retrieval-augmented methods improve factual accuracy by grounding language models in external knowledge, but retrieving over unstructured text often introduces irrelevant context and offers limited control over the retrieved information. Structured knowledge bases offer a more controllable alternative, yet they are expensive to construct and often brittle to reason over. To address these limitations, we propose KBevo: a co-evolving framework that jointly learns to construct a structured knowledge base and reason over it for knowledge-intensive question answering. By optimizing both components end-to-end with QA outcome rewards, our method enables reasoning success to directly improve the quality of the constructed knowledge base. This leads to larger, better-connected knowledge structures with higher answer reachability, while also improving compositional factual reasoning and controllability compared to standard retrieval baselines.

Volume-based accuracy rewards retrieval-augmented generation (RAG) systems for guessing: a system that answers everything outscores one that declines when its knowledge base cannot support an answer. Building on the confidence-target analysis of Kalai et al. (2025), we present a penalty-aware evaluation framework for deployed RAG products, combining (i) asymmetric scoring (correct +1, wrong -4, abstain 0), (ii) knowledge-gap canaries, questions whose answers are verifiably absent from the knowledge base, so that any answer constitutes ungrounded generation from parametric memory, and (iii) a failure-attribution pipeline that separates retrieval, generation, and abstention-policy failures. Applying the framework to three commercial RAG systems and a no-retrieval baseline on SimpleQA-Verified (1,000 questions x 3 repeats, graded blind by a cross-family three-judge panel with 98.9% unanimity), we find that accuracy when answering is closely clustered across systems (97.0-98.0%), while canary violation rates differ roughly sixfold (16.7% vs. 98.1%). The systems are separated less by what they answer correctly than by whether they answer at all when they should not, and penalty-aware scoring reorders the volume-based ranking accordingly; the reordering is stable across penalty settings from k=1 to k=9. All code, configurations, transcripts, and judge votes are released for independent audit.

Autonomous robots performing laboratory tasks depend on 3D reconstruction pipelines that can turn raw camera streams into actionable object representations within the latency budget of a physical control loop. Neural 3D reconstruction methods have demonstrated high-quality view synthesis, but their real-time viability across the compute platforms on which laboratory robots actually run remains poorly characterized. In this work, we present a systematic compute-platform benchmark of neural 3D reconstruction methods, evaluating NeRF and 3D Gaussian Splatting training and rendering on GPU-enabled computing devices ranging from single-board computers to server-class nodes, and place Meta's SAM3D single-image reconstruction on the same axes to quantify its latency and fidelity gap relative to per-scene optimization. Our results show that Gaussian Splatting yields higher rendering quality than NeRF at greater GPU cost, and that onboard compute is insufficient for full per-scene optimization at interactive rates. Our preliminary assessment on SAM3D indicates that it delivers plausible object geometry within seconds, but with detail mismatches that can compromise downstream manipulation. Together, these findings motivate tiered pipelines in which lightweight feed-forward reconstruction sustains the real-time perception-and-tracking loop for laboratory robots, while heavier neural reconstruction is scheduled selectively on suitable compute.

Retrieval evaluation for retrieval-augmented generation (RAG) is increasingly designed around whether retrieved passages contain evidence that can support generation, rather than topical relevance alone. We study whether this closer alignment with downstream evidence needs also makes retrieval evaluation more useful for the decisions built from it. Across five retrieval benchmarks and an end-to-end TREC RAG 2025 setting, we examine an answer-support signal in four roles: comparing retrievers, guiding retrieval training and system selection, predicting downstream answer quality, and filtering the evidence supplied to a generator. The signal changes retrieval rankings, but its downstream value is not uniform. It does not reliably improve retriever training; the benefit of using it for system selection depends on how the generator is instructed to use the retrieved evidence; and retrieval scores based on it do not robustly predict answer quality on unseen topics. In a direct evidence intervention, human annotators confirm that filtering preferentially preserves passages containing useful answer evidence, yet different answer evaluators reach different conclusions about whether the resulting answers improve. These results show that making retrieval evaluation more closely reflect the evidence needed for generation does not by itself make every downstream use of that evaluation more reliable. RAG evaluation methods should therefore be assessed with respect to the particular comparisons, decisions, and conclusions they are intended to support.

Diffusion language models (DLMs) generate text by iteratively denoising masked sequences, but standard decoding either fixes the sequence length or relies on ad hoc stopping rules, often leading to unnecessary denoising steps. We recast length selection as a discrete-time survival problem over the end-of-sequence token and propose a plug-in, training-free length predictor that can be added to any existing DLM. Across reasoning and code-generation benchmarks, survival-guided length decoding speeds up inference by up to 7 times while preserving task accuracy. We further find that predicted lengths vary widely even within the same dataset, making model performance sensitive to the chosen length.

Large language models are increasingly deployed as autonomous agents serving users on behalf of companies, placing them in settings where user and deployer interests can conflict. When an agent knows that a user is owed something its deployer would prefer to deny, does it remain honest? Answering this is difficult because false statements can reflect either ignorance or hallucination rather than deception. To address this challenge, we introduce KnownLieBench , a knowledge-verified benchmark that first confirms through a neutral probe that an agent knows a user's entitlement, and then evaluates whether it makes false claims once an incentive to deny that entitlement is introduced. Specifically, KnownLieBench covers eight customer-service domains and 112 grounded cases, conducts multi-round dialogues with a trust-tracking customer agent, and separates deception emerging from incentive alone from deception produced under explicit instruction. Across eighteen proprietary and open-weight models, emergent deception varies substantially across model families and domains. We further use the benchmark for post-training, finding that honesty-directed fine-tuning reduces deception under incentive, while deception-graded fine-tuning increases lie success on honest-control dialogues without increasing lie frequency under incentive. By verifying entitlement knowledge before scoring deceptive behavior, KnownLieBench reduces the confound between lying and not knowing and enables more rigorous auditing and steering of agent honesty.

Tool-augmented LLMs must arbitrate between two fallible sources when a tool return conflicts with their parametric memory, yet existing evaluations measure source preference without establishing source correctness. We introduce MemToC, a controlled benchmark for post-tool-return arbitration with executable tools. MemToC comprises 6,504 evaluation episodes constructed from 542 quality-controlled factual questions, independently elicited model-specific closed-book answers, and controlled tool returns of known correctness. These components instantiate four source-correctness cases; tool-error and no-tool conditions are separate controls. Across five open-weight 7-9B models, tool returns strongly dominate elicited closed-book answers. The four instruction-tuned models retain a verified-correct answer against an incorrect tool in only 6.5-17.1% of eligible cases, follow a correct tool in 86.0-93.1%, and repeat the tool return in 78.4-86.0% of cases where both sources are wrong. No cross-model ordering remains stable across three instruction-wording variants with the question and episode content held fixed. We compare prompting with SFT and DPO using chain-level cross-fitting over ToolHop, so questions sharing an underlying fact never straddle training and evaluation. We apply an asymmetric success criterion: correct-answer retention must improve without a detected reduction in correct-tool following. SFT and DPO meet this criterion on the same two of four instruction-tuned backbones. Improvements rarely come cleanly: 19 of 20 tested method-model combinations reduce abstention after tool errors or on unanswerable inputs. Transfer beyond MemToC is positive but partial and depends on the model and presentation frame. Correctness-conditioned arbitration can be improved through fine-tuning, but gains must be evaluated jointly with correct tool use, abstention, and robustness to formulation.

Large Language Models (LLMs) increasingly act as autonomous agents executing complex, long-running procedural skills. Existing agent runtimes maintain execution by continually appending observations, actions, and intermediate reasoning traces to an ever-growing conversation history, causing latency degradation and context-poisoning failures over long horizons. We present SKILL.state, a runtime architecture that replaces append-only conversational history with an explicit, mutable execution state. At each execution step, the model receives only the immutable skill specification, the current structured execution state, and the latest observation. Intermediate reasoning is discarded immediately after producing a validated state update, preventing prompt growth with execution history. Across diverse datasets, models, and execution environments, SKILL.state improves task accuracy while substantially reducing cumulative token consumption. Our results demonstrate that explicit execution state is an effective and architecture-agnostic abstraction for scalable long-horizon agent skills.

Native visual reasoning treats visual generation as the medium of reasoning itself: visual states (i.e. images and videos) are not merely inputs to be understood or outputs to be rendered, but first-class substrates for problem solving beyond language. Yet progress remains bottlenecked by the lack of scalable training tasks, reliable feedback, and controlled comparisons across generative substrates. In this work, we introduce VBVR-Pro, a closed-loop testbed that makes native visual reasoning through generation trainable, verifiable, optimizable, and experimentally controllable. 1) Task scaling. VBVR-Pro turns visual reasoning into a controlled task space of 300 procedurally generated tasks. Models trained on VBVR-Pro show strong transfer beyond the proposed suite across seven external visual reasoning benchmarks such as RISE-Video, MME-CoF-Pro, and BabyVision. 2) Verifiable rewards. VBVR-Pro provides verifiable reward scorers for task-grounded evaluation. Through a systematic study of leading MLLMs as judges, we identify recurring failure modes of the prevalent VLM-as-a-judge paradigm. In contrast, the proposed scorers are grounded in deterministic, task-specific rules, achieve fine-grained alignment with human judgments. Importantly, they serve as reliable reward signals for large-scale multi-task reinforcement learning and demonstrate stronger post-RL performance across visual reasoning tasks. 3) Mechanism study. VBVR-Pro enables controlled modality studies across more than 30 image, video, and interleaved generators. Our analysis shows that video generation remains strongest for tasks requiring persistent spatiotemporal state tracking, while interleaved generation provides a compute-efficient alternative. Critically, ablations and probing suggest the presence of vision-native trajectories that are crucial to visual reasoning. We release all data, models, scorers, and code.

Zero-shot cross-task generalization, where a policy must execute manipulation tasks never seen during training, remains a central challenge in robot learning. In large language models, a novel task can be performed simply by specifying it in the context, without any parameter update. This form of in-context learning (ICL) turns generalization into a problem of task specification. To achieve cross-task generalization, we bring this paradigm to robotic manipulation, and argue that the natural task specification for manipulation is a human video: unlike language, it provides rich visual cues about the intended task evolution. We present Zero-WAM, a causal video-action model that executes unseen tasks by following in-context human video guidance. To address the scarcity of task-rich paired human-robot data, we propose an automatic pipeline that converts task-sampled robot trajectories into semantically matched human videos, yielding HumanGen, a dataset of 74.2K human-robot ICL pairs across 8.6K tasks. For model training, we further introduce an in-context future chunk prediction (IFP) objective that suppresses shortcuts learned from seen tasks and forces the policy to draw task information from the video prompt. On seven unseen tasks in RoboTwin 2.0 simulation, Zero-WAM achieves a 47.0% average success rate, an absolute improvement of 29.5 percentage points over the strongest video-action baseline. In real-world evaluations, it follows human video guidance to generalize to unseen task configurations involving multi-object scenes, long-horizon manipulation, and fine-grained insertion.

Generative world models provide robots with predictive models of how the world evolves under interaction, with growing potential for simulation, planning, policy evaluation, and robot learning. Beyond clip-level future prediction, a unified generative formulation should relate actions to consequences, support flexible horizons and continuous interaction, and enable reward-driven optimization. We introduce WALL-SS, a world model that generates visual futures through Scale-wise autoregressive Scaling, enabling action-controllable and long-horizon robotic simulation. WALL-SS represents embodied trajectories as causal sequences of temporally interleaved observations and actions, making action-dependent state transitions explicit while naturally supporting variable-length generation, streaming extension through reusable causal states, and direct optimization through sequence probabilities. To make this formulation effective over long horizons, we generate each future observation in a coarse-to-fine manner and develop three complementary components within the same hierarchy. Action-conditioned next-scale prediction injects scale-aligned action representations to improve action-future coupling and model both successful and failed behaviors. Scale-compressed long-horizon memory retains recent interactions at fine resolution while compressing distant observations and actions, with scale-wise dream forcing enhancing robustness to self-generated context. Finally, on-policy alignment optimizes autoregressive visual dynamics with action-following and long-term consistency rewards while preserving the pretrained visual distribution. Experiments show that WALL-SS improves action following and trajectory accuracy, supports coherent minute-long streaming rollout under bounded memory, and consistently benefits from on-policy alignment in reducing action drift and long-horizon inconsistency.

Designing machine learning algorithms for wireless resource management is labour-intensive: the architecture, the loss function and the training recipe are all specified by hand. We demonstrate that this design layer can be surrendered to an autonomous agent in its entirety. We adopt the autoresearch protocol, in which an AI coding agent edits a training script, runs a fixed-budget experiment, and retains or discards the change according to a single immutable metric. We grant the agent authority over the architecture family, the input representation, the output parameterization, the loss function and the task-sampling law, and set it a target chosen for its difficulty: sum-least-percentile-rate power control across a multicell network. The formulation targets cell-edge throughput and is non-convex, non-smooth and strongly NP-hard away from its max-min vertex. Safeguards render the results trustworthy: a hash-pinned evaluator, an enforced inference contract and a pre-registered falsifier per experiment. In eighty-one unattended experiments over twenty-six hours, the agent reached $99.5\%$ of a converged minorization-maximization reference in one fixed-cost inference pass, at roughly $600\times$ lower inference cost, closing $94\%$ of the gap from its first working architecture, with one parameter set serving every network size and percentile target. It recovered provable structure rather than tuned constants: the output parameterization it discovered reproduces the exact max-min-optimal allocation at the minimum percentile, for every value of the trained weights.

Grasp synthesis is a core task in robotic manipulation, for which the solution typically forms a multimodal distribution rather than a point estimate. Generative robotic grasping aims to learn this distribution with deep generative models such as diffusion and flow-based approaches. The iterative nature of such generative models makes them flexible and generalizable; however, multi-step sampling impedes the time-critical operation required in robotics. We devise an approach to fast generative grasping based on MeanFlow on the product Lie group $\mathcal{G} = \mathrm{SO}(3) \times \mathbb{R}^3$. The training objective couples a purely algebraic semigroup consistency condition with Riemannian Conditional Flow Matching on $\mathcal{G}$ that anchors the average velocity to the data distribution. The resulting Lie Group-constrained MeanFlow formulation samples reliable grasps in $\leq 5$ network evaluations, matching the grasp generation performance of state-of-the-art diffusion and flow-based models on the ACRONYM dataset at millisecond-scale inference latency (up to $39\times$ speed-up). We further demonstrate that the approach directly translates to real-world robotic grasping without additional training or domain adaptation, exhibiting robust grasp synthesis under observation noise.

Scaling generalist vision-language-action (VLA) policies is severely bottlenecked by the inherent heterogeneity of embodied data, which spans diverse robot morphologies, camera configurations, and low-level action spaces. Existing paradigms typically address this mismatch through explicit action retargeting, human-to-robot video synthesis, or dataset-specific adaptation branches, fundamentally hindering the joint learning of a unified policy. We introduce UCAG-P, a camera-centric unified action formulation that structurally aligns heterogeneous embodied datasets into a shared geometric action space. Rather than treating robot-specific commands as the shared policy target, UCAG-P represents manipulation through camera-observable anchor motion in image and camera-frame coordinates, treating robot arms, humanoids, and human hands as different embodiments of a common action schema. A geometry-conditioned action translator combines predicted motion with target-embodiment kinematics to produce executable controls. The resulting decoupled architecture allows a shared VLA policy to learn transferable manipulation geometry while retaining embodiment-specific controllability. UCAG-P is trained on 4.03K hours of robot and simulation data and 2.34K hours of human demonstrations. A single checkpoint reaches 98.3% on LIBERO, 88.7% and 89.2% on RoboTwin Easy and Hard, 82.0% zero-shot on LIBERO-Plus, and 62.0% on RoboCasa GR-1, without benchmark-specific fine-tuning.

Reasoning in language allows foundation models to spend more test-time compute on hard problems, such as those requiring decomposition, constraint tracking, and prediction of future consequences. Whether this mechanism can improve robotic manipulation remains unclear, where long-horizon tasks require tracking partial progress, reasoning about object relations, recovering from mistakes, and steering noisy low-level policies. In this paper, we study whether VLMs can be trained to reason directly in natural language to guide low-level manipulation policies. We introduce $R^3$, a simple post-training recipe that turns off-the-shelf VLMs into robotic reasoners: it first mid-trains a VLM on expert-generated reasoning traces to initialize the desired reasoning style, then improves the reasoner with single-step rubric-based RL from offline action data. Unlike prior robotic reasoning methods that mostly use structured traces as auxiliary supervision, $R^3$ trains free-form language reasoning to produce test-time guidance for action. We instantiate $R^3$ on Language Table and simulated bimanual grocery packing, two controlled testbeds for studying robotic reasoning and long-horizon manipulation. $R^3$ improves exploration and generalization across unseen tasks and significantly outperforms instruction-only imitation learning baselines on both benchmarks. Our analyses suggest that free-form language reasoning can function as a test-time compute mechanism for steering low-level policies. Our project page is available at https://robotic-reasoner.github.io/.

Frequency encoded photonic qubits promise a scalable path towards high-dimensional quantum information processing, but require efficient components for coherently mixing frequency modes. Coupled cavity modulators provide this functionality by using only a single driving microwave tone to couple hybridized optical supermodes. Here, we demonstrate a heterogeneously integrated thin-film lithium-niobate-on-silicon coupled-cavity modulator that realizes tunable bidirectional frequency mode transformations, including \(50/50\) beam splitting and near complete frequency swapping with \(>20~\mathrm{dB}\) pump extinction at a \(10~\mathrm{GHz}\) supermode splitting. Because the electro-optic film is bonded onto a foundry fabricated silicon photonics platform, the approach is compatible with co-integration of photon pair sources, spectral filters, active tuning elements, and single photon detectors. We also bond thin-film lithium tantalate onto the same coupled-cavity platform, demonstrating material flexibility for scalable integrated frequency bin quantum photonic circuits.

Advanced automation is a key technology for enhancing the safety of ground operations amidst the increasing density of commercial air traffic. While motion forecasting is a well-studied task in autonomous driving, its application to airport surface movements remains underexplored. To enable efficient and accurate prediction in this domain, we propose DESCENT, a transformer-based architecture designed to handle heterogeneous dynamics and strict topological constraints. Our approach features a Potential Reachable Set (PRS) context sampling mechanism that adaptively collects airfield environment context across diverse operational phases. Combined with a detection transformer-based decoder, DESCENT generates accurate trajectory forecasts. Extensive evaluations on the Amelia-10 benchmark demonstrate significant performance improvements over state-of-the-art baselines. These gains are especially pronounced in safety-critical scenarios, where our domain-aware sampling provides critical long-horizon context necessary for safe navigation.

Multi-agent large language model (LLM) workflows have emerged as a powerful paradigm for solving complex, open-ended tasks through collaborative reasoning among specialized LLM agents, but they incur substantial operating costs due to repeated LLM invocations and long-horizon context accumulation. Existing cascade routing methods make one-shot, query-level decisions and cannot adapt to the dynamic, state-dependent nature of multi-step workflows, in which the right LLM at each step depends on evolving task progress, remaining task difficulty, and cost-efficiency requirements. We present ProgRouter, an online progress-guided routing framework that adaptively selects LLM agents across workflow steps to preserve task-solving quality while adhering to time and cost budgets. ProgRouter introduces a multi-view task progress scorer that combines coarse workflow outcome regimes with fine-grained signals on subtask completion, progress trends, and workflow state quality. Then, a dual-path task progress predictor and an adaptive meta-gating mechanism estimate the progress gain for each candidate routed LLM. ProgRouter makes online step-wise routing decisions that balance progress gain, task time budgets, and long-term operating cost efficiency. Experiments on HumanEval Plus, MBPP, MATH-500, and ASQA, spanning agentic code generation, mathematical reasoning, and retrieval-augmented long-form question answering, demonstrate that ProgRouter reduces the operating cost relative to key baselines while maintaining strong task-solving performance.

Digital and energy technologies depend on microelectromechanical and power electronic components whose performance is critically impacted by rapid, cyclic deformations. Real-time information on their operation has remained inaccessible due to the need for nanosecond and nanometer resolution in fully integrated devices. We break this limitation by imaging the complete switching cycle of an industrial GaN high electron mobility transistor through stroboscopic dark field X ray microscopy at a fourth-generation synchrotron, resolving electromechanical and thermal micro strain fields across the entire device and correlating them with time dependent voltage characteristics. Coupled simulations benchmarked against the measurements reproduce electric field evolution and transient thermal hotspots. This combined approach provides direct insight into device physics and informs design strategies for next generation energy and information processing technologies.

Autonomous R\&D agents now write, run, and improve executable artifacts under automated evaluation---but largely as laboratory instruments: shown on curated benchmarks, with gains that are hard to trace to a cause and costs well above what sustained engineering practice absorbs. The limitation is structural. Most systems treat each attempt as nearly self-contained, so logs, memories, and search trees record what happened without establishing which design element produced an improvement, whether its evidence survived validation, or how it recombines with others. Long campaigns therefore keep re-learning the same lessons. We introduce Praxist, a lineage-centered generational system that converts reproducible artifacts and evaluator outcomes into a typed evidence graph of findings, lane-structured frontiers, and agendas. Separating local artifact construction from cohort-level evidence synthesis lets later attempts inherit validated mechanisms, unresolved claims, and useful constraints, and leaves results attached to an inspectable lineage. On the standardized 75-task MLE-bench suite, the finalized official-grader results give Praxist 60 medals (80.0\%), 49 of them gold, against 55 medals (73.3\%) and 34 gold for a Claude Code baseline on Claude Opus 4.8---at a recorded model spend of US\$3,054 versus US\$38,370, roughly a twelfth of the cost. Four case studies---quantitative trading, LiDAR-inertial-visual SLAM, tokamak magnetic control, and rocket landing---carry the same process into open-ended engineering problems, improving on each task-native baseline in headline accuracy, survival, or resource cost, with the discovery path on record. Stronger artifacts at an order of magnitude less spend, each backed by an auditable lineage, are, to our knowledge, first brought together here: the operating profile production research requires, not the one a benchmark demonstration establishes.

Neighborhood livability is commonly assessed with static built-environment indicators, such as facility proximity, street connectivity, and access to public space. These measures describe available opportunities but do not directly represent how residents with different mobility capacities, household roles, schedules, and care responsibilities experience the neighborhood. This paper presents a prototype framework that uses a spatial knowledge graph (KG) and large language models (LLMs) to generate and revise household schedules, followed by rule-based feasibility checking and GIS-based network materialization. The spatial KG integrates residents, residences, facilities, neighborhood context, and sampled road hubs; Graph-RAG retrieves each household's nearby spatial context, including candidate POIs and approximate walking times, for the scheduling LLM. The LLM produces structured household schedules, while rules are used for lightweight repairs and auditable feasibility checks. The LLM then revises schedules in response to identified feasibility issues. A routing module derives the actual travel paths, travel times, modes, and event histories from the road network. The resulting events support synthetic resident-agent interviews about daily convenience, travel burden, activity feasibility, and household coordination. A prototype demonstration in a Shenzhen neighborhood shows that nominal facility availability does not necessarily imply convenient access: residents with limited mobility and households with care responsibilities experience greater travel and coordination burdens. The framework offers an auditable way to connect spatial opportunity, household activity constraints, and resident-specific livability interpretation, while keeping simulated experience distinct from observed perception.

Physical AI models are evaluated on suites of benchmarks that differ across model reports, leaving the model-by-benchmark matrix sparse and the relationship between benchmarks unmeasured. We construct a matrix of 51 models on 12 physical AI benchmarks, selected from a registry of 51 benchmarks and 152 models by reporting density, combining scores from model cards and benchmark papers with our own evaluation runs under each benchmark's official protocol. We measure how much information the benchmarks share and show quantitative evidence of Redundancy. Redundancy affects reported rankings: collapsing the two substitute pairs into single columns moves 22 of 51 models by three or more places under an equally weighted average. We then select benchmarks greedily under a utility combining score dispersion with variance not explained by the already-selected set, and obtain a four-benchmark subset retaining 78.5\% of the utility of all 12, on which we fit a Bradley--Terry ranking. The procedure requires only benchmark-level scores with sufficient overlap and is not specific to physical AI.

Multi-agent systems (MAS) sometimes already have the potential to answer correctly, but still report a wrong answer. Explaining this outcome is difficult because generation, communication and final answer-selection rules usually change simultaneously. We conceptualize multi-agent reasoning as an evolutionary pipeline of candidate generation, peer communication and terminal selection, wherein consensus without quality control can exhibit patterns of memetic drift. We study two questions: (1) when an LLM judge provides effective selection pressure by supplying a signal of answer correctness for candidates generated in a multi-agent system, and (2) when using that signal improves the reported answer. To map judge reliability, we analysed 15,336 questions from MMLU-Pro, GPQA, MedXpertQA and MuSR, with Humanity's Last Exam analysed separately. To test these rules, we replayed 81,390 fixed candidate pools drawn from 16,278 questions across five benchmarks. We report three findings. (1) A correct answer is often already present among the generated candidates, but the system can still converge on and report a wrong answer. (2) Judge reliability is not a fixed trait of the model, but varies with the task, the generator and how rare the correct answer is. (3) Combining answer frequency with the judge's evaluation changed only the final answer-selection rule and raised accuracy from 63.82% to 70.82-70.95%, primarily by rescuing correct answers that were outnumbered by popular errors. In the systems studied here, the value of generating more candidates depends on whether those extra samples make correct answers present, frequent or recognisable. By isolating generation, recognition and selection, these findings establish a diagnostic basis for designing multi-agent architectures that protect generated correct answers from being lost.

Agentic systems enable the intelligent automation of complex workflows, specific to additive manufacturing this is applicable for complex tasks such as process parameter optimization for mechanical properties. This work investigates the AI enabled agentic process optimization within Selective Laser Sintering (SLS) to iteratively improve the tensile and flexural properties of 3 different materials on the Inova Mk1. These materials include PA12 GF, PA11 Onyx, and PA12 Blend (volume mixture of 25% PA12 GF and 75% PA12 White) and with using knowledge from previous builds and minimal guidance from the user, the agentic system was able to optimize process parameters over a small number of iterations to achieve comparable TDS specified mechanical properties. This work showcases the ability for an agentic system to continually learn from updated data, enabling the intelligent automation of complex tasks such as process parameter optimization for selective laser sintering.

Autonomous agents increasingly perform bounded software tasks under an orchestrator that retries, resumes, and budgets them. The machinery such orchestrators reach for is the service mesh's: retry, timeout, and error-rate circuit breaking. We report a failure study of a production agentic software-delivery platform over 147 numbered incidents spanning 81 runs, each with a measured cost and, in most cases, a mutation proof reproducing the failure. All three assumptions those primitives rest on are violated in practice, and we quantify the consequences: a loop of fifty-four consecutive successful tool calls no error-rate breaker could see; a progress signal constant by construction, guaranteeing a false trip on the third repair round and driving one run from six of six components to three; twenty-one events accumulated across six invocations of one delegation, making a correct, idempotent component unwinnable; a misrouted failure that woke five components for a two-component fault, leaving three bystanders regressing working code; and twelve incidents in which the enforcement layer blocked correct work, the most expensive costing 107 agent turns and zero accepted writes. We find one cross-cutting cause and its dual. Identity adequacy: in five separate subsystems an identity that failed to discriminate produced a confident wrong answer, and two of them derived the corrective rule independently. Evidence adequacy: a reliability decision may be taken only on evidence capable of moving, attributable to what it measures, and deterministic under identical conditions. From the findings we derive seven reliability primitives whose enforcement unit is the delegation rather than the message, and specify the controlled evaluation the study motivates but does not constitute.

Terahertz (THz) photonics offers a promising platform for next-generation 6G wireless communications, on-chip spectrometers, and non-invasive biomedical sensors. In the near- and mid-infrared, monolithically integrated photodetectors are a cornerstone of photonic integrated circuits (PICs). In the THz range, however, such integration on low-loss dielectric waveguides remains an open challenge: to date, room-temperature on-waveguide detection has been realized only through hybrid flip-chip assembly of discrete elements, which leads to poor mode coupling, larger footprint, and reduced mechanical stability. Here we address this gap by demonstrating a room-temperature graphene sub-THz detector monolithically integrated on a high-resistivity silicon dielectric waveguide for D-band (110-170 GHz) operation. We couple an hBN-encapsulated graphene channel to the guided mode using a tapered slot-line antenna patterned directly onto the silicon surface. The detector achieves a voltage responsivity of 11.5 V/W and a 3 dB bandwidth of 1.94 GHz, with the latter currently limited by parasitic inductance of the readout interconnects rather than by the intrinsic graphene response. This platform provides a practical pathway to fully integrated room-temperature THz photonic circuits, where electrostatic gating and impedance matching can improve both responsivity and bandwidth by roughly an order of magnitude.

This work tackles the problem of scalable perception for robotic packing with low-cost, low-resolution depth sensing. We propose a framework where reconstruction cues drive next-view selection and grasp evidence updates a per-object stability estimate, jointly deciding what to acquire next and when to grasp. During the reconstruction, a low-resolution Next Best View (NBV) strategy explicitly avoids redundant views while preserving task-relevant geometry. We validate the approach in two steps: (i) an ablation study of the utility function under very low resolution, and (ii) a full end-to-end evaluation across policies, showing how low-resolution perception is a practical, scalable option for robotic packing.

Contact-rich manipulation requires precise interaction feedback. While vision-centric imitation learning is prevalent, external visual observations provide indirect and ambiguous cues about contact states, particularly under occlusion or subtle object--gripper interactions; dedicated tactile or force sensors can provide rich contact information but introduce additional hardware complexity, calibration requirements, and deployment costs. To bridge this gap, we propose VISTA-Policy, an imitation learning paradigm that utilizes the Visual Deformation Field (VDF), a 3D displacement representation of a compliant gripper, as high-dimensional visuo-physical feedback. The framework integrates: 1) a Physics-Aware Encoding Engine for real-time VDF decoding; 2) an Energy Aggregation Denoising Mechanism to isolate true interaction signals; and 3) a Deformation-Augmented Policy Network with incremental gripper actions for precise closed-loop correction. Extensive evaluations on Cross-Scale Object Grasping, Cap Unscrewing, and Calligraphy Writing demonstrate that VISTA-Policy outperforms the strong pure-vision baseline 3D Diffusion Policy and the tactile baseline. VISTA-Policy further demonstrates substantial out-of-distribution generalization to unseen object scales and robustness against dynamic disturbances, offering a durable and cost-effective route toward general-purpose fine-grained manipulation in unstructured environments. Project videos and supplementary materials are available at: https://sites.google.com/view/vista-policy.

Multi-arm collaboration is becoming a core capability in embodied manipulation. Recent vision-language-action (VLA) models integrate perception, language, and control, but most represent language as a single global instruction and do not provide an explicit mechanism for assigning and composing arm-specific behaviors. This design limits transfer to collaboration patterns that differ from those observed during training. We present MA-VLA, a unified framework for multi-arm collaboration via atomic action assignment. MA-VLA decomposes cooperative behavior into mid-level atomic prompts and allocates them to individual arms, enabling explicit subgoal specification and compositional reuse across tasks. To reduce reliance on fixed execution roles, we introduce Arm Shuffle, a training-time permutation of the observation, state, and assigned atomic prompts for each arm. This permutation enforces role-agnostic instruction following and supports recomposition into unseen coordination patterns, which we term multi-arm compositional generalization. We also construct a benchmark in which test-time collaboration patterns are absent in training set. Across simulation and real-world evaluations, prior state-of-the-art VLAs largely fail under these unseen collaborations, while MA-VLA consistently succeeds. These results indicate that structured, per-arm atomic action assignment offers a practical route to scalable generalization in multi-arm embodied systems. Code, models, and data are available at https://github.com/zhangzaibin/future-robots

Global Tensor Motion Planning (GTMP) solves motion planning with batched tensor operations over a layered multipartite graph. We generalize GTMP so that adjacent-layer edges are realized by any black-box local planner (e.g., linear interpolation, splines, sampling-based planning, trajectory optimization, or generative sampling). We provide two anytime policies on top of this generalization: Anytime GTMP with random restarts at a fixed budget, which covers every homotopy class almost surely, and AO-GTMP with informed expansion with growing budgets, which converges to the optimal cost. We prove that a single sampled graph covers every endpoint-fixed homotopy class admitting a \(δ\)-clear representative of bounded length. We also prove that additional samples per layer reduce the per-layer miss probability exponentially, whereas stronger local planners reduce the required layer count only sublinearly. On manipulation benchmarks the method matches state-of-the-art performance, and on 2D navigation it returns batches of topologically diverse solutions, while the informed baselines concentrate on one or two classes.

Contact-rich manipulation requires adapting to contact states that can evolve substantially within an action horizon. However, chunk-based vision-language-action models predict complete action chunks from observations collected before execution, leaving tactile conditioning stale during execution. Existing tactile-reactive approaches typically rely on separate high-frequency controllers, which increase both architectural and training complexity. In this paper, we introduce TacForcing, a streaming action-generation framework that effectively incorporates execution-time tactile feedback. Instead of employing a separate reactive controller, TacForcing replaces the standard action expert with a streaming action expert to generate actions conditioned on the evolving tactile observations acquired during execution. TacForcing also introduces Execution-Aware Tactile Attention (EATA), which restricts tactile conditioning to actions nearing execution, thereby reducing the temporal mismatch between tactile acquisition and action execution. Across six simulated UniVTAC tasks and three real-world contact-rich manipulation tasks, TacForcing achieves average success rates of 65% and 69%, respectively, outperforming strong baselines in both settings.

Large language models (LLMs) have enabled AI scientist systems to automate scientific discovery, yet existing approaches most rely on static prompting or fixed workflows and fail to accumulate experience for continual improvement. We propose HypoForge, an experience-guided multi-agent framework that learns reusable scientific skills for automated hypothesis generation and hypothesis testing. HypoForge is built on the observation that these two stages involve different supervision signals. For hypothesis generation, where explicit feedback is unavailable, HypoForge adopts an adversarial generator--discriminator mechanism to improve reasoning through comparative critique. For hypothesis testing, where empirical feedback is available, HypoForge learns testing skills from execution outcomes and ground-truth results. By matching skill learning strategies with stage-specific supervision, HypoForge enables continual improvement without fine-tuning foundation models. Experiments on hypothesis generation and testing benchmarks show that HypoForge consistently outperforms existing AI scientist frameworks and skill-level variants. Further analysis demonstrates the effectiveness of the proposed stage-specific skill learning paradigms.

Generalist vision--language--action (VLA) policies learn long-horizon behavior mainly through short-horizon action prediction and reveal little beyond sampled commands. This creates two coupled bottlenecks: a single action target must implicitly absorb task progress, intermediate intent, and local reliability, while these control states remain hidden during execution. Inspired by functional principles of biological sensorimotor control, we introduce LM-X , which organizes prediction across task, event, and motor scales without claiming anatomical correspondence. Three explicitly supervised signals are emitted online and directly condition action generation: return-to-go (RTG) measures visible task progress, event-to-go (ETG) identifies the next semantic transition, and heteroscedastic action flow estimates local reliability through propagated variance. Explanation is therefore intrinsic to control rather than generated post hoc. Before a costly 20-day pretraining run on 64 NVIDIA B200 GPUs, a controlled five-task pretraining gate verifies the design: the complete model improves success by 16.0 points over the action-only backbone and by 10.8 points over the strongest single-head variant. We then train LM-X on more than 20,000 hours of real-robot trajectories, including over 1,000 hours of failed policy rollouts. LM-X achieves 74.1\% across 50 randomized-hard RoboTwin2.0 tasks versus 55.4\% for GR00T N1.7, and 68.6\% versus 50.7\% across seven real-robot tasks. RTG tracks semantic progress and visible regression, while variance rises during hesitation and oscillatory control. These results show that explicit multi-timescale predictive state can strengthen control while exposing interpretable internal estimates.

As generative AI gains traction, researchers are investigating its potential to serve as proxies for humans. From undergoing cognitive psychology experiments to experiencing an epidemic, generative agents, agents powered by generative AI models, produce realistic human behavior when prompted. This study explores the sensitivity of these generative agents' behavior to prompt modifications and varied persona names of the agents. To assess this sensitivity, we use a generative agent epidemic model, wherein each agent is prompted daily on whether it wants to isolate or commingle with other agents. We found that using synonymous prompts results in negligible changes to the model's outcomes. However, minor variations in prompts, as well as contextual changes, do influence the model's results. Lastly, our data indicates that different persona names assigned to generative agents, specifically those imbued with personas, do not significantly impact epidemic outcomes.

The variability of photovoltaic (PV) generation poses significant challenges to the reliable and efficient operation of grid-connected microgrids. Accurate PV output power forecasting and efficient energy scheduling strategies are essential not only for optimizing PV system operation but also for improving the overall performance and reliability of the system. This study proposes a long short-term memory (LSTM)-based PV power forecasting model integrated with a multi-objective scheduling framework for a grid-connected PV-battery energy storage system (BESS). The proposed approach enables detailed performance monitoring and assessment by quantifying how PV forecast accuracy influences key operational metrics, including PV self-consumption ratio, grid energy cost, grid injection, and battery utilization. Three forecasting scenarios (perfect forecast, persistence model, and LSTM-based forecast) are compared to evaluate their impact on system performance and operational reliability. Results show that the LSTM-based forecast reduces root mean squared error (RMSE) by 6% compared with the persistence model, increases the PV self-consumption ratio from 78.1% to 84.5%, and reduces grid injections by 82%. The analysis also highlights trade-offs, as higher battery throughput associated with improved performance may contribute to accelerated aging. These findings demonstrate the importance of accurate PV forecasting in improving system performance and ensuring reliable operation. Future work will focus on probabilistic forecasting to properly quantify uncertainties, incorporate load prediction, and develop smart control strategies that allow grid-support functionalities from the PV side.

Bimanual manipulation in cluttered, contact-rich environments remains challenging because it requires coordinated motion generation, interaction-aware planning, and reliable execution under tight kinematic constraints. We present PRISM, a projection-integrated sampling-based Model Predictive Control (MPC) framework that uses a GPU-accelerated physics simulator as an online world model for complex dual-arm manipulation. The main algorithmic contribution is a QP-guided control sampling strategy that decouples trajectory exploration from kinematic feasibility. At each MPC step, sampled joint-velocity trajectories are projected onto the set of motions satisfying joint position, velocity, acceleration, and jerk bounds, together with an initial-velocity boundary condition, before rollout evaluation. This enables broad yet feasible exploration of coordinated bimanual behaviors. To support efficient online execution, we derive a custom ADMM/Bregman-splitting QP solver that exploits joint-wise separability and reusable matrix factorizations. We further use Bayesian optimization to tune task-cost weights offline, reducing manual parameter selection. We evaluate PRISM on challenging variants of PerAct$^{2}$ tasks, including obstacle-constrained ball transport, tray transport, cube handover, and box lifting. Experiments show improved robustness and task success relative to representative sampling-based baselines, while maintaining real-time or near-real-time execution. We also demonstrate successful sim-to-real transfer on dual UR5e manipulators, highlighting the practical potential of physics-based online planning for contact-rich bimanual manipulation. Project details, including code and supplementary videos, are available at \href{https://sites.google.com/view/prismbimanual}{\texttt{https://sites.google.com/view/prismbimanual}}.

Vision-Language-Action (VLA) policies have advanced language-conditioned robotic manipulation, yet action-imitation objectives provide only weak supervision for metric 3D structure and short-horizon physical evolution. Geometry-enhanced policies mainly improve current-scene grounding, whereas predictive policies often model future dynamics in RGB or latent spaces and may incur substantial deployment cost. GaussianDream demonstrates that training-time current Gaussian reconstruction and future Gaussian prediction provide effective 3D supervision, but its dense VGGT/TGE-based prefix jointly carries state, dynamics, and action-conditioning information. We present \textbf{\methodname}, a compact, policy-native extension that inserts \textbf{World State Tokens} and \textbf{World Prediction Tokens} directly into the VLA backbone. A training-only \textbf{World Representation Head} decodes these tokens into a Current World and coupled Future Prediction over shared Gaussian primitives, while static--dynamic factorization preserves persistent structure and focuses residual motion on interaction-relevant regions. At inference, the head, renderer, auxiliary objectives, and VGGT/TGE pathway are removed, leaving only 20 world tokens without online Gaussian decoding or rollout. \method achieves \textbf{98.6\%} on LIBERO and \textbf{87.8\%} on LIBERO-Plus, with clear gains under Camera and Layout shifts. Real-robot experiments further improve average success from 29.2\% to 52.5\% over reproduced $π_{0.5}$ while maintaining efficient closed-loop control.

Image-goal navigation using lightweight topological maps is a practical paradigm for indoor robot deployment: the map requires only geotagged images, and localization relies on visual matching rather than precise pose estimation. However, learned waypoint predictors can produce targets that violate geometric constraints or deviate from the global path. Executing these waypoints safely further requires a local planner capable of collision avoidance, yet existing systems either lack one or rely on fixed parameters that cannot adapt to confined spaces. To address these limitations while retaining the navigational intuition of the learned predictor, we present EgoNav, a hierarchical system that implements this idea by generating candidates from semantically segmented traversable regions and scoring them alongside the learned waypoint for geometric safety, directional coherence, and fidelity to the learned prior. An adaptive local path planner then executes the refined waypoint with parameters modulated based on the refinement outcome. Experiments in Habitat-sim and on a physical humanoid robot show that EgoNav consistently outperforms contemporary baselines in both success rate and path efficiency.

Traditional task-and-motion planning (TAMP) approaches primarily focus on defining sequences of actions along with the necessary geometric and kinematic constraints to execute long-horizon tasks. However, their applicability in real-world settings is limited, as they typically assume simplified object models that overlook key physical properties critical for the successful execution of contact-rich tasks. Moreover, they often use sub-symbolic reasoning during motion planning, which drastically increases planning time and decreases overall success rates. We propose a method that leverages a TAMP approach, defining object-centric abstractions of execution constraints, called Unified TAMP (U-TAMP), to execute robotic tasks involving interactions among objects with heterogeneous shapes, sizes, and materials. Using a Vision-Language Model (VLM), we generate abstractions of inter-object affordances for characterizing physical interaction constraints between objects in contact-rich tasks, such as grasp and support constraints. These constraints are used to enrich the U-TAMP planning domain to deal with objects with variable physical properties. We perform experiments in simulated kitchen table organization scenarios and compare our results with those of the original U-TAMP, as well as a state-of-the-art VLM-based planner that leverages common sense knowledge of objects' affordances for plan generation. Our approach achieves significantly higher planning success rates and improves planning times by one to two orders of magnitude compared to other methods.

Robot policies are predominantly learned with classical parametric variants of imitation learning or RL, where training stores the agent's behavior exclusively in the policy's network parameters, putting a heavy burden on the representation learning algorithm. We propose a new navigation agent equipped with non-parametric memory which explicitly indexes prior steps leading to critical events. The advantages are twofold: first, it allows the policy to outsource some of its behavior into an explicit memory; second, it encourages a form of continual learning by allowing an agent to collect data from its testing episodes during deployment and therefore to better generalize to OOD situations. In the context of social navigation, we show that this improves the agent's capability to retain sparse, high-cost failures, such as human collisions. If the policy is trained in simulation, this also naturally addresses the sim-to-real gap, partially, by basing some of the decision making on real data. We integrate the explicit memory into a recurrent PPO architecture and use hidden states for memory retrieval to capture continuous spatiotemporal dynamics. The goal of exploiting rare, high-impact events is achieved by leveraging the RL agent's advantage signals. We train our agent in simulation with a combination of photorealistic rendering and non-visual crowd simulation and show that the agent is robust with respect to OOD social behavior.

Vision-Language-Action (VLA) models provide a versatile foundation for general robotic manipulation, yet they exhibit significant brittleness when confronted with novel task distributions. While In-Context Imitation Learning (ICIL) offers a training-free alternative, existing frameworks suffer from an adaptation bottleneck that hinders the effective translation of expert context to executable actions. This failure originates from superficial retrieval mechanisms and an inherent behavioral inertia that anchors the policy to its pre-trained priors. To address these limitations, we present RA-VLA, a retrieval-augmented VLA framework that integrates behavior-aligned context retrieval with a grounded execution pipeline. By enforcing faithful adherence to functional cues within a scalable architecture, RA-VLA facilitates seamless task adaptation while preserving inference efficiency. Our empirical evaluations across the LIBERO benchmark and a real-world UR5e environment demonstrate that RA-VLA achieves superior success rates and computational efficiency, establishing a robust framework for training-free robotic adaptation.

Action-conditioned world models have become an important foundation for embodied prediction, planning, and synthetic data generation, but their errors under new task and scene distributions are often concentrated in localized spatiotemporal regions such as robot arms, manipulated objects, contact areas, and occluded objects. This paper presents ConfAL-WM, a confidence-guided active learning framework for post-training embodied world models. Built upon EVAC, we attach a lightweight confidence probe to UNet decoder features and predict dense confidence maps in the latent space. These maps are aggregated into task-, frame-, and patch-level scores, enabling both efficient data selection and localized training enhancement. Our pipeline first retrains the confidence probe and warms up EVAC with a small subset of target-domain data, then performs task-level prescreening to allocate sampling budgets, and finally applies selected-data retraining with optional frame or patch weighted data enhancement. Experiments on RoboTwin2.0 show that confidence-guided selection improves post-training efficiency, while dense frame and patch weighting further enhances prediction quality and embodied trajectory consistency compared with scalar reward, progress, and judge-based scoring baselines. A quick visual overview of this work is available at https://ConfAL-WM.github.io.

Hardware kernel optimization requires repeated compilation, correctness testing, profiling, and revision. LLM agents can automate parts of this process, and stronger foundation models, longer context windows, and longer execution horizons have improved optimization within individual tasks. These advances alone do not enable an agent to learn from completed optimization runs. Existing kernel-optimization agents seldom preserve a decision, its observed execution feedback, and the later decisions that use that evidence. Retaining every prior trajectory is also impractical because an expanding history competes with the current task for context. We present KOPE, an experience-driven framework for hardware kernel optimization. KOPE records optimization trajectories with correctness and performance feedback in Experience Graph Memory, then uses Active Context Management and Injection to retrieve relevant experience under a fixed token budget. The graph retains decision order, observed outcomes, and alternative branches, allowing evidence collected on the target hardware to inform later optimization steps and tasks. Under the same GLM-5.2 setting, the geometric mean of KOPE's per-operator speedups is $1.54\times$ that of CANNBot, the strongest competing baseline. In a complete 53-operator ablation, Active Context Management and Injection raises pass rate from 60.0\% to 84.6\%, increases the evaluator-reported positive-field geometric mean from 0.0382 to 0.0661, and reduces optimization token consumption from 15.9B to 1.113B tokens relative to passive agent-led context construction. Enabling Experience Graph Memory raises full-suite pass rate from 55.2\% to 84.6\% and yields a $1.43\times$ geometric-mean speedup on valid timing comparisons. These results support continual optimization through external experience while the foundation model remains fixed.

This paper studies resource allocation for MapReduce-based collaborative computing over heterogeneous wireless devices powered by renewable energy harvesting. We formulate a long-run average throughput maximization problem that jointly optimizes computing load, phase time allocations, transmit power, and per-device energy consumption, subject to battery evolution, CPU frequency, and latency constraints. To solve this problem online without prior knowledge of channel states or energy arrivals, we propose a DDPG-CVX algorithm that couples Deep Deterministic Policy Gradient (DDPG) with convex programming. DDPG determines the per-slot energy budget for each device from observed battery and channel states; the remaining resource allocation variables are then resolved to global optimality by an embedded convex solver. This two-phase decomposition reduces the action-space dimensionality of DDPG while preserving per-slot solution quality. Simulations show that DDPG-CVX achieves 1.25$\times$$\sim$32.36$\times$ the throughput of representative benchmarks.

To apply the techniques of embodied artificial intelligence to human-oid robots for complex manipulations, dexterous robotic hands are indispensable, which are restricted by the dexterity and tactile perception capability. In this work, we proposed a novel design of tendon-driven five-fingered hand with dis-tributed tactile perception. With a soft-rigid-hybrid structure employed, both compliance and operational force are endowed to the hand. Dual-modality tactile sensing elements are distributed on the distal and middle phalanges of all five fingers, enabling the simultaneous detection of static contact and dynamic force variations. Manipulation experiments, including counting gestures, finger-to-thumb pinching, object grasping, and bottle-grasp tactile recording, demonstrate the feasibility of the integrated actuation-perception system.

This paper presents a novel two-phase maskable proximal policy optimization (TP-MPPO) algorithm, which maximizes the system goodput counting request throughput with strict service level objective (SLO) compliance for large language model (LLM) inference services in wireless edge networks. In the first phase of TP-MPPO, we optimize the task offloading decisions by MPPO with action masking mechanism, effectively avoiding exploring invalid actions and reducing the action space. In the second phase, closed-form solutions are derived for uplink bandwidth allocation; a greedy algorithm is designed for downlink bandwidth allocation to provide immediate rewards for the MPPO in the next round. The two stages alternate till convergence. Simulation results demonstrate that TP-MPPO can improve the system reward by 33.3%--87.5% compared to its benchmarks and achieve the highest goodput.

This paper investigates joint beamforming and port selection in multi-cell fluid antenna-assisted (FAS) networks. In such networks, active beamforming and discrete FA port selection are coupled through intra-cell and inter-cell interference and are jointly optimized to maximize the weighted sum-rate (WSR). We develop a federated representation learning (FedRep) framework with a position-aware dual-branch deep neural network (PA-DNN). The PA-DNN uses channel state information and port positional encoding as inputs, and jointly outputs beamforming vectors and port selections through two task-specific branches. To support decentralized training across heterogeneous cells, the FedRep framework shares global beamforming-related parameters among base stations while keeping port-selection parameters local for cell-specific adaptation. Simulation results show that the proposed scheme achieves a higher weighted sum-rate than conventional FL and port-selection benchmark schemes.

Despite their strong ability to generate code, large language models often fail to produce secure code, as their outputs frequently contain security vulnerabilities. Secure code generation is inherently challenging because it requires solving a multi-objective problem: functional correctness and security. Existing approaches address this challenge by injecting external security knowledge or by using agentic feedback and iterative refinement. However, guideline retrieval often leaves the generator to translate generic advice into task-specific secure implementations, while shared-dialogue multi-agent feedback can blur role boundaries and suffer from context bloat. We present MACGen, a multi-agent framework that integrates planning, security analysis, code synthesis and refinement to jointly optimize security and functionality. A planner constructs a step-by-step plan to satisfy functional requirements. A security advisor identifies likely CWEs and synthesizes task-specific guidelines, a coder then generates code grounded in these artifacts, and a reviewer issues perspective-separated feedback. Rather than sharing full dialogue histories, each agent receives only structured artifacts from upstream stages, enforcing role specialization and reducing uncontrolled context growth. On CWEval and BaxBench, MACGen improves F&S@1 over direct prompting by 19.61 and 10.57 percentage points (pp) on average, respectively.

Beyond-visual-range (BVR) air combat is a challenging reinforcement-learning domain characterized by partial observability, long-horizon decision making, energy management, and limited weapons. We present BVR Sim, an open-source Gymnasium-style environment designed for heterogeneous air-combat reinforcement learning. BVR Sim supports multiple JSBSim aircraft models, including the F-15, F-16, F/A-18, and F-22, with configurable weapons, sensors, controllers, and opponents. A unified tactical action interface specifies desired heading, altitude, speed, and weapon release above aircraft-specific inner-loop controllers, enabling policies to operate across heterogeneous platforms. The environment provides interchangeable Python and accelerated C++ backends, entity-oriented observations, compositional rewards, scripted opponents, replay and visualization, and adapters for multi-agent learning frameworks. At a 0.4-s decision interval, the C++ backend achieves 104 simulated seconds per wall-clock second in 1-vs-1 and remains practical through 10-vs-10 scenarios. A policy trained only on the F-16 transfers without retraining to four unseen aircraft, reaching a 45.5% mean win rate with aircraft-specific controller adaptation. MAPPO and HAPPO experiments further verify end-to-end compatibility with standard multi-agent reinforcement-learning pipelines.

Real-world humanoid tasks involve physical interaction with objects and humans, yet current controllers either reject external forces as disturbances or restrict compliance to limited body links while ignoring angular effects. We present LAC, a general whole-body controller that simultaneously realizes commanded Linear and Angular Compliance for wrenches applied to the upper body. First, we synthesize whole-body compliant responses into a large-scale augmented dataset. Sampled force and couple events are imposed on contact frames extracted from human interaction data. At each contact link, the external force and a virtual torque from the passively yielding kinematic chain drive a virtual admittance under the commanded stiffness. Subsequently, teacher-student reinforcement learning trains a single policy to track the compliant motions under external wrenches. Finally, extensive simulation and real-world experiments demonstrate whole-body compliant responses to wrenches across the upper body, monotonic modulation over the full range of both stiffness commands, and applicability to teleoperated loco-manipulation tasks. Project website: https://lac-humanoid.github.io/

Recent vision-language-action models offer a path toward robots with broader repertoires than conventional task-specific systems. Construction deployment, however, requires a precise inventory of worker activities and the capabilities needed to execute them. We present TARCAT, an occupation-grounded taxonomy derived from 91 O*NET tasks across seven high-employment construction occupations and 30 instructional videos of physical work. TARCAT defines 41 action primitives in 12 groups and three classes and provides a mechanism for composing parameterized primitive sequences into reusable skills. This human-interpretable structure can organize demonstrations, specify robot requirements, and support coding agents that retrieve and extend skill libraries. We also demonstrate selected primitives on a DOBOT CR3 arm with a CRAFT hand. TARCAT thereby provides a common vocabulary for analyzing human work and developing general-purpose construction robots. Annotations are available at https://github.com/AICPS/TARCAT-Taxonomy.

In this paper, we propose RAEM, a robust autonomous exploration framework for quadruped robots operating in multi-floor environments. Most existing ground-robot exploration approaches rely on planar traversability representations, which cannot adequately represent the overlapping structures and cross-floor connectivity of multi-floor buildings. Although tomography-based representations provide effective traversability modeling for multi-floor navigation, maintaining a global tomography map incurs substantial computational overhead for online exploration with frequent replanning. Moreover, sparse and fragmented LiDAR observations in stairwells can degrade local traversability estimation, leading to irregular viewpoint placement and temporary topological disconnections. To address these challenges, RAEM adopts a hybrid local-global traversability representation, in which a local tomography map and an explicitly categorized local 3D grid map are used for online terrain analysis and connectivity evaluation, while an elevation-aware global topological graph is incrementally constructed from these local spatial representations for efficient cross-floor exploration planning. We further introduce a staircase center alignment strategy to reduce abrupt yaw variations during climbing and a dual path searching mechanism to recover guidance paths when the global topology is locally disconnected. Extensive simulation and real-world experiments demonstrate robust and computationally stable autonomous exploration across multi-floor structures, including continuous exploration of a five-floor stairwell.

Vision-language models (VLMs) have emerged as a powerful source of supervision for reinforcement learning, enabling agents to leverage rich semantic knowledge during training. Inspired by the success of preference-based reward learning (PbRL) in reinforcement learning from human feedback (RLHF), vision-language model generated image-based preferences provide an effective source for learning reward functions. This can be done by visually comparing two outcomes through the Bradley-Terry (BT) model. However, this pairwise formulation utilizes only two observations at a time, despite VLMs being capable of ranking multiple candidates. The Plackett-Luce (PL) formulation can shape a reward model with listwise rankings as opposed to pairwise preferences, allowing for a more suited use of a VLM based ranking. In this work, to our knowledge, we introduce the first framework that combines VLM-generated preferences with the Plackett-Luce model for reward learning. We evaluate our approach on Meta-World manipulation tasks and show that Plackett-Luce (PL) reward models can train robotic policies from VLM-generated rankings as effectively as pairwise Bradley-Terry, $K$-wise Bradley-Terry, and RL-VLM-F baselines. Across all environments, at least one PL ranking size ($K \in \{3,4,5\}$) consistently performs with or outperforms other methods in mean success rate. Unlike pairwise methods, which are restricted to $K=2$, PL supports different ranking sizes and can therefore be adapted to the environment and desired feedback format. Our best PL configuration achieves an 86% mean final success rate and matches the Oracle baseline on Drawer Open. Overall, these results demonstrate that listwise VLM preference supervision is a competitive and flexible approach to reward learning for reinforcement learning.

This paper develops a certified constraint-reduction method for distributed model predictive control with tube-tightened exponential control barrier functions (eCBFs) in multi-agent systems. At each prediction stage, pairwise agent--agent and agent--obstacle eCBF conditions define halfspaces in the local control space. Rather than enforcing all such halfspaces, a geometry-adaptive subset is retained and a Farkas certificate verifies that the reduced admissible set is contained in the full tightened set. For planar inputs, cone coverage is characterized through the largest angular gap: two extreme directions suffice in the strict half-plane regime, while other geometries initialize with three retained constraints and escalate only when certification fails. Conic multipliers and nominal-aware offsets are obtained in closed form, without an auxiliary optimization, and the resulting construction preserves any nominal control already admissible for the full tightened set. Consequently, the reduced controller inherits the robust safety guarantee of the underlying tube-eCBF formulation. In a ten-follower, four-obstacle study, the method retained fewer safety constraints on average, reproduced the full filter's nominal accept/reject decisions with no true safety violations, and achieved increasing computational gains as the constraint count and prediction horizon grew.

Paradigmatic interaction models explain how collective behaviors can emerge in complex systems from interactions among the constituent agents. In bio-inspired swarms, however, interactions alone may not suffice to bring the population to a desired aggregate configuration within a prescribed time horizon, as needed in applications ranging from targeted therapy to collective transport and emergency evacuation. In the present work, we consider finite-horizon minimum-energy collective steering for inertial swarms that are subject to stochastic disturbances. We focus on the mean-field representations of these multi-agent systems driven by Cucker--Smale alignment or Morse attraction--repulsion interactions. Our objective is to steer the swarm between prescribed endpoint distributions using a state-feedback control, where the endpoint specifications can be full phase-space distributions (positions and velocities) or position marginals alone. Our formalism is rooted in the theory of Schrödinger bridges, which has inspired contemporary developments spanning statistical inference, biological modeling, stochastic control, and generative learning. Within the bridges framework, the uncontrolled interacting stochastic dynamics are viewed as a prior model, and the optimal control as the minimum-energy corrective drift needed to realize the prescribed distributions. We derive nonlinear, coupled necessary optimality systems with a time-symmetric structure reminiscent of classical Schrödinger bridges, and propose nested fixed-point schemes to numerically solve them. Numerical examples show that the obtained optimal control (corrective drift) can dynamically exploit or counteract the interaction forces, depending on whether the latter are favorable or adversarial to the steering task.

Large language models (LLMs) frequently exhibit \emph{sycophancy}: they adapt their answers to a user's stated beliefs or preferences instead of reporting what they hold to be true, which lowers factual accuracy and can amplify misinformation. This paper proposes a methodology for mitigating sycophancy that employs the Bayesian Truth Serum (BTS), a peer-prediction mechanism, as the reward in Group Relative Policy Optimization (GRPO) to fine-tune an LLM. BTS pays an answer for being \emph{surprisingly common}, that is, more frequent among respondents than those respondents themselves predicted. We treat a group of responses from a model for one question as those respondents, so the reward is a function of the model's own outputs and fine-tuning needs neither labels nor preference annotations. We prove that in the large-group limit a sycophantic response earns strictly lower expected reward than an honest one. We also prove that if the entire group agrees in advance on a symmetric answering rule, it cannot earn a higher information score than under truthful reporting. On our true/false benchmark the reference model's answer-flip rate under user pressure decreases from 23% to 4%, and its accuracy under that pressure increases from 80% to 93%. Our reward outperforms SMART and is comparable to synthetic-data fine-tuning and to pinpoint tuning, all three of which train on labels. It spends considerably more compute in exchange, which makes it suitable when labeled data is scarce. Peer Truth Serum, which also pays a premium for a rare answer but elicits no prediction report, reproduces the effect. A peer-prediction reward computed inside a single GRPO group therefore reduces sycophancy without labels, and comparing mechanisms suggests that the premium paid for a rarer answer drives the effect.

In dynamic collaborative systems, the selection of reliable collaborators is critical to ensuring effective task execution. Existing trust evaluation methods often rely on unidimensional or scalar representations, which fail to faithfully capture a collaborator's true trustworthiness, thereby motivating a shift toward multi-dimensional trust modeling. However, due to the asynchrony of collected trust-related data across different dimensions, as well as the complex intra- and inter-dimensional dependencies embedded within these data, multi-dimensional trust evaluation remains challenging. To address these challenges, we propose TrustFormer, a task-specific multi-dimensional trust evaluation framework. Specifically, TrustFormer leverages task identifiers and device-generated timestamps to synchronize heterogeneous trust-related data across historical collaborations. It further employs cross-temporal and cross-dimensional attention mechanisms to jointly model temporal dynamics and inter-dimensional correlations, thereby effectively learning the multi-dimensional trust evolution of potential collaborators from historical performance data. In addition, according to the multi-dimensional resource requirements of tasks, potential collaborators' multi-dimensional resource trust is evaluated. Finally, by synthesizing these multi-dimensional trust profiles, the framework enables the optimal collaborator selection. Experimental results demonstrate that TrustFormer outperforms existing methods by yielding a 40.8% improvement in trust evaluation accuracy and enabling more reliable collaborator selection.

We introduce CIPS, a contract-driven execution abstraction for managing computational latency and sampled-data updates in safety-critical cyber-physical systems (CPS). A fundamental challenge in real-time control is that physical safety certificates are defined spatially, yet predicting their validity under non-zero computation and handoff latency requires online numerical integration of plant dynamics. CIPS resolves this operational dichotomy by systematically compiling heterogeneous spatial safety certificates into normalized, unit-rate temporal contracts entirely offline. This transformation abstracts complex plant dynamics, exposing a deterministic, O(1) temporal budget to a generic sampled-data scheduler. We formally prove that this architecture guarantees global hybrid safety invariance under bounded computational latency and asynchronous execution. Finally, we validate the framework via an autonomous vehicle braking benchmark, demonstrating a 60.5 times reduction in micro-architectural evaluation overhead compared to latency-aware event-triggered control (LA-ETC) while preserving safety bounds.

The impairment of the hands can seriously affect the abilities of every individual to perform the every-day activity, so the design of stable and controllable support devices is a significant field of study. This paper is about the design and implementation of an automated bionic hand which is dedicated to the coordinated finger movement through the simplified and efficient actuation mechanism. The method that the proposed system was designed on is the tendon-based method whereby the servo motors generate the movement of the fingers, with assistance of the angular control which is calibrated. An actuation is controlled by a microcontroller that will be programmed by use of an Arduino-based microcontroller to carry out programmed gestures that include open hand, fist, pinch and half flexion. It has an interface that is voice command enabled to make it easy to interact with a Bluetooth based sender receiver architecture which offers an option of executing trained commands which are immediately converted to finger actions. To explore the motions behavior, finger coordination and control response to the input, the behavior of the experiment system is tested. The actuation of the fingers was found to take a total of about 7-8 seconds to achieve full flexion of all fingers in a sequence. The system showed repetitive and constant motion throughout several actuation cycles without loss of any apparent tension or precision of control. There was a stable grasp of objects of different shapes and sizes, which implied consistent coordination between the fingers. These findings indicate that the proposed system offers predictable and steady control behavior and has a simple and efficient mechanical and control architecture.

Natural language driven autonomous co-scientist workflows involve a fundamental trade-off between flexibility and reasoning at the expense of determinism, reproducibility, and observability. Such agents increasingly must communicate across institutional boundaries, where federation topology can shape latency and cost. We systematically evaluated these tradeoffs using a controlled ablation on a production agentic platform for science. We use a verifiable task: given a protein sequence, we ask an agent to confidently characterize its function by routing across common tools. We compare federation topology, classic RL vs LLM-driven harnesses, language model, and prompt expertise. We also stratify results by protein novelty. We find that the choice of LLM dominated prediction quality far more than topology or prompting (Opus ~92%-94% vs o4-mini ~40%-50%). The PPO policy was nearly as accurate as the best LLM (88%) at zero token cost, fastest latency, and perfect consistency, but yields no reasoning trace. Expert prompted LLMs reached the highest accuracy but were high-cost and less consistent; prompt dependence was largest when the task was hardest. Federation imposed a negligible penalty on performance. These results offer actionable guidance for deploying agents for scientific workflows: for routine, verifiable tasks, a cheap deterministic policy delivers near-frontier accuracy with complete reproducibility, while flexible LLM reasoning is best reserved for open-ended discovery.

We introduce CRESSim-Neo, a batched GPU simulation engine for surgical robotics and robot learning. CRESSim-Neo combines position-based simulation of rigid bodies, deformable tissues, fluids, and strands with batched rendering, surgery-specific sensing, and a GPU-resident data pipeline. The engine supports applications including tissue manipulation, fluid suction, suturing, cable-driven robots, and ultrasound image synthesis. Direct access to physics and rendering buffers enables GPU-resident robot learning and zero-copy PyTorch integration using DLPack. We demonstrate CRESSim-Neo across rigid-body, deformable-body, and fluid simulation tasks, including vision-based and surgical robot-learning scenarios. On an NVIDIA RTX 4090, the engine achieves up to 2.03 million environment steps per second for 8192 parallel CartPole environments, and scales to batched surgical scenarios involving tissue deformation, fluid interaction, and ultrasound sensing. Overall, CRESSim-Neo provides a unified and scalable platform for surgical simulation, synthetic data generation, and surgical robot learning.