Physical AI Brief

Coding agents have emerged as a promising paradigm for robot manipulation: a language model writes the robot controller as a program, and agents built in this way now operate robots without robot-specific training.Whether this paradigm is also safe, however, has not been asked. We evaluate coding agent under a safety constraint, where each task pairs a manipulation goal with an obstacle the robot must not touch. The agent pursues the goal but collides with the obstacle in most cases, treating task completion as its sole objective while neglecting safety. The agent reasons about the obstacle in its traces, and the prompt already forbids touching it, so neither perception nor instruction is at fault; the fault lies in the planning, where the stated constraint never becomes a priority. By decomposing manipulation into a route phase and a contact-rich moment, we locate the source of the failure. Along the route, the model cannot prioritize the safety constraint, having no notion of a clearing route and none of replanning once a chosen route becomes infeasible. At the contact, it is unaware that contact execution is bounded by the same constraint. To close this gap, we present SafeHarness, which equips the model with two obstacle-aware harnesses that enable it to prioritize the safety constraint. Obstacle-aware route planning grounds the objects as bounding boxes and draws candidate routes over them as sequences of waypoints. The agent then plans a route in advance, verifies it, replans when necessary, and only then executes it. Obstacle-aware contact execution instead selects the contact position so that the contact itself avoids the obstacle. SafeHarness attains 71.9% task success and 87.5% collision avoidance, surpassing the previous SOTA by 6.5% and 27.0%, respectively. These results are $2.3\times$ and $1.5\times$ those of the same agent without harnesses.

Complex robotic manipulation tasks frequently require a long-term memory of past events and actions. As conditioning on full histories renders policies prone to spurious correlations and degrades performance, many approaches to policy memory involve compressing historical information through expensive VLM queries in-the-loop to process only task-salient information. In this paper, we propose an alternative approach in which computationally intensive VLM queries are made during train-time to learn a lightweight latent memory that can be efficiently queried at deployment time. Our representation, which we call the \textbf{workspace token}, is trained by (1) using a VLM to identify current and historical information necessary for completing a task, then (2) distilling these into the workspace token using a set-reconstruction decoder loss. In both simulation and hardware, we show that the workspace token can be used as a drop-in replacement for observations during deployment, enabling policies to solve memory-intensive tasks without the need for VLM reasoning in-the-loop. Interestingly, we found that workspace tokens are not only more lightweight but also lead to better policy performance.

Perceiving and remembering the visual world is fundamental to navigating and interacting with our environment. Current 4D foundation models, such as camera-controllable video models or 4D reconstruction models, can perceive and reconstruct dynamic environments, but how well they remember what they have perceived remains an open question. Existing benchmarks largely rely on pixel-level metrics and lack ground truth for objects once they leave the field of view, making them unable to evaluate visual memory in an object-centric manner against references. To fill this gap, we introduce PersistBench, a dataset and metric suite that leverages 360° videos as omniscient ground truth and proposes three evaluation aspects: object permanence, motion continuity, and appearance preservation. Evaluating various models across diverse categories reveals that current models can only maintain short-term consistency that degrades significantly once objects leave the field of view. Our findings highlight the gap between current model capabilities and robust visual memory ("seeing is not remembering"), providing guidance for future development of 4D foundation models. Dataset and code are available on the project page: https://guangzhaohe.com/persistbench.

A splash lives for a fraction of a second: sheets tear into ligaments and droplets, appearance is view-dependent and nearly textureless, and little persists long enough to track. Reconstruction research has consequently focused on smoke, synthetic liquids, or gently deforming surfaces. To our knowledge, no synchronized multi-view dataset of splashing liquids exists. We therefore introduce a benchmark of 20 real scenes, from coherent streams to violent splashes, captured by seven synchronized, calibrated 4K cameras at 60 fps, with manually refined per-view liquid and container masks and fixed evaluation splits. We further present SplashSplat, built on a single principle: impose physical structure only where the observations can constrain it. Per-frame liquid SDFs fused from the masks provide the geometry, level-set transport between consecutive SDFs yields a coarse velocity field, and Lagrangian carriers advected along this flow, corrected against each new observation and reseeded where coverage is lost, decode local Gaussians for differentiable rendering. SplashSplat outperforms state-of-the-art dynamic Gaussian splatting methods on our real captures and on a synthetic benchmark, with physically more plausible motion and a lower training cost. The same representation supports temporal interpolation and style transfer without re-optimization.

Professional design requires any-color control: the ability to specify an object's target color with any 24-bit hex value for image generation and editing. Prior work has explored color generation, editing, and colorization, but often relies on dedicated color representations or specialized inference procedures. Advances in large language models offer a simpler starting point: even compact models can associate hex values with color semantics. We present Paint-Anything, which learns a shared hex-prompt interface for generation and editing through object-level color supervision. We develop a data pipeline that constructs Paint-500K from real images through object grounding, perceptual color labeling, and editing-pair synthesis. Since shadows make real-image labels only approximate colors, we complement this supervision with pure-color anchors whose pixels exactly match their paired hex values. These anchors are used only at high-noise timesteps, leaving low-noise training to natural images. We further introduce Any Color Benchmark (ACBench), comprising ACBench-T2I and ACBench-Edit, to measure object-level hex color fidelity across both tasks. On FLUX.2-4B, Paint-Anything improves ACBench-T2I and ACBench-Edit scores by 85.3% and 28.3%, respectively, relative to the base model, with ablations supporting the training recipe. It also achieves the highest average CompColor score among the compared methods.

Frontier coding agents are increasingly trusted to work autonomously for long periods, yet an agent's final response is often the only account of that work a user sees. We quantify the propensity of frontier agents to \emph{overclaim} task completion, a misrepresentation that can mislead the user. An agent overclaims when its final response contradicts information in its context. This definition requires no inference about intent and is independent of task success. We introduce \emph{OverclaimBench}, an evaluation suite composed of five file-review scenarios, transcript-based coverage measurements, and registered planted defects. We evaluate eight proprietary frontier models in their own production command-line interfaces, and four open-weight models under a single fixed harness on OverclaimBench and find that 1) agents do not read all the files they were asked to review in 67.9\% of runs; 2) among runs where not all files are read, agents are \emph{misleading} 80.4\% of the time (59--96\% per model), either falsely claiming to have read all files or omitting that coverage is incomplete; 3) requiring delegation to subagents increased reading coverage, but among reviews that remained incomplete, a large majority were still misleading; and 4) agents that falsely claimed a complete review missed planted defects at about 1.8 times the rate of agents that read every file, showing that claims of completion can conceal substantive failures. Together, these results show that agents' final responses are not reliable accounts of their actions.

Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering performance, yet existing work typically evaluates harnesses as monolithic systems, leaving the effectiveness of individual components unclear. To enable component-level comparisons, we study this question with a lightweight coding harness whose execution loop is fixed while three components are varied: planning, action space, and context management. Across four models evaluated on SWE-Bench Verified and Terminal-Bench 2.1, we evaluate 176 matched settings spanning five context-management strategies, four context-window budgets, and targeted ablations of planning and action space. We find that: (1) Context management becomes increasingly valuable as the context-window budget tightens, with most of its benefit coming from preventing context-overflow failures. (2) Staging rule-based elision before LLM-based summarization provides the strongest overall efficiency among the context-management strategies, whereas making elided content recoverable adds machinery that models rarely use and yields no accuracy gain. (3) Planning shifts from an accuracy scaffold for weaker models to a cost saver for stronger models, with little change in accuracy. (4) Predefined tools improve performance for models with weaker bash proficiency, whereas bash-capable models can operate effectively with a bash-only interface and achieve substantially lower cost, especially on command-line-centric tasks. Trajectory-level analysis explains these effects: context management extends execution trajectories without substantially altering agent behavior, planning changes where trajectories stop, and the action space changes the granularity at which code is written. These findings inform model- and budget-aware harness design and provide a modular framework for evaluating future harness components.

World modeling enables intelligence to anticipate consequences, guide interventions, and learn from interaction. Yet predictive models remain domain-specific: can a common learning principle support world modeling across radically different systems? We introduce JEPA-Anything, a domain-agnostic framework based on orthogonal predictive factorization (OPF). Extending joint-embedding predictive architectures, OPF decomposes latent targets into complementary factors, learns them through dedicated pathways, and recombines them within a shared predictive design. We evaluate JEPA-Anything across seven domains: vision, biology, clinical trajectories, control, molecular dynamics, physical fields, and weather. Experiments span representation learning, intervention prediction, out-of-distribution generalization, and long-horizon dynamics, including 10 matched dynamics tasks, forecasting of over 1,000 clinical events, and 100-step molecular rollouts across four systems. Against matched JEPA baselines, JEPA-Anything improves reported metrics on all 10 dynamics tasks and reduces single-intervention prediction error on Interventional Pong by 34.8%. It achieves the lowest one-step and 100-step molecular errors among compared methods in all four systems. Beyond prediction, a factor-nominated biological intervention receives experimental support in cell co-cultures, patient-derived organoids, tumor fragments, and mice; latent orbital modes recover the Keplerian scaling exponent with a fitted slope of -1.4991. These results support a common factorized predictive principle across heterogeneous worlds, connecting world modeling with intervention and experimentally grounded scientific discovery. Code: https://github.com/Gen-Verse/JEPA-Anything

Generative models are increasingly used to solve scientific inverse problems, but existing evaluations still focus primarily on whether a method can produce a single plausible reconstruction. This is insufficient for ill-posed problems, where multiple solutions may be consistent with the same sparse or noisy observations. In these settings, a method can achieve strong pointwise accuracy while still failing to capture the true posterior through mode collapse, overconfident uncertainty, or averaging incompatible solutions. We introduce PosteriorBench, a benchmark for evaluating the distributional accuracy of generative inverse solvers. PosteriorBench evaluates four physics-based inverse problems: Darcy flow inversion, Poisson source recovery, carbon capture and storage, and light transport material inference. For each task, we construct high-fidelity reference posteriors using computationally heavy but established procedures such as rejection sampling and Markov chain Monte Carlo, enabling direct assessment of whether solvers recover the full set of solutions rather than the single best sample. We pair these references with a five-metric posterior evaluation suite: posterior-mean error, posterior-standard-deviation error, maximum mean discrepancy, sliced Wasserstein distance, and radially averaged power-spectrum error. These metrics assess pointwise accuracy, marginal uncertainty, distributional alignment, and global frequency fidelity. The benchmark spans sparse sensing, low-resolution observations, nonlinear forward models, varying noise levels, and multimodal priors, with a unified pipeline for distribution matching and uncertainty quantification. Our experiments reveal substantial distribution-matching gaps across current solvers, while showing that neural operators improve resolution robustness, and guidance weights and generation noise are key to posterior-variance calibration.

Hierarchical planning frameworks combine skills from multiple robot control policies for long-horizon task execution, where determining when to terminate the current skill and advance to the next subtask is essential. Existing approaches often rely on pre-designed completion signal checkers that are hard to obtain in real-world execution. Large-scale vision-language models (VLMs) offer strong reasoning capabilities, but their decision boundaries are not inherently aligned with task completion criteria, while cloud deployment and lengthy reasoning introduce substantial latency, limiting real-time monitoring. We propose StageGuard, an agentic distillation framework for accurate and efficient stage-transition decisions. StageGuard combines teacher-model reasoning with demonstration trajectories to generate structured explanations of subtask completion and policy switching. A lightweight student VLM uses these explanations to generate compact self-explanations, which are used for supervised fine-tuning. We evaluate stage-transition prediction on trajectories from two benchmarks and assess closed-loop task success through integration into hierarchical robot control on BEHAVIOR-1K, with further validation on real robots. Results show substantial improvements in stage-transition prediction while supporting efficient online monitoring.

Multi-turn agents trained with reinforcement learning (RL) receive a single scalar reward per trajectory, which motivates self on-policy distillation (OPD) to supply dense token-level supervision from a self-teacher with privileged task skills, letting a skill-free student internalize them. This recipe, however, is undermined by two findings in agentic tasks: privileged information alone does not always make a teacher reliable, and the benefit of teacher supervision is stage-dependent. We therefore propose RetireOPD (Self-Retiring On-Policy Distillation), which first optimizes a decoupled, skill-conditioned teacher with environment rewards and then trains a skill-free student jointly with RL and OPD. Rather than following a predefined distillation schedule, RetireOPD adopts Adaptive Retirement: the student drops the teacher on its own once their discrepancy stops shrinking and it reaches a target fraction of the teacher's success rate, after which training proceeds with RL alone. Across Qwen2.5 models from 1.5B to 7B, RetireOPD improves ALFWorld success rate over RL baseline by 14.1% to 18.8% and WebShop accuracy by 11.8% to 19.0%, and surpasses its own skill-conditioned teacher in every setting.

Action chunking is widely used for action generation and execution in Vision-Language-Action (VLA) policies, yet existing approaches commonly use a fixed action horizon. During a rollout, different task stages may require different levels of action continuity, control precision, and closed-loop feedback, making a fixed horizon unable to accommodate changing control requirements. We propose \textbf{GeoAAC}, a geometry-based adaptive action chunking method for flow-based VLA policies that adjusts the action horizon according to the reliability of the current action prediction. We show that the geometry of Flow Matching denoising trajectories provides process-level information for characterizing prediction reliability, with geometric variation across action prefixes remaining positively correlated with predictive uncertainty. GeoAAC uses this prefix-wise geometry to construct a horizon-wise geometric profile and adaptively determine the action horizon from a single generation without additional training. Experiments with GR00T N1.5 and π0.5 on LIBERO, LIBERO-Pro, RoboCasa365, and real-world manipulation tasks show consistent improvements over fixed-action-horizon baselines and existing adaptive methods, including up to 8.7 percentage points in simulation and an increase in average real-world success rate from 53.3\% to 74.4\%.

Generative agents are increasingly used to select and narrate video highlights, but they typically operate over unstructured or frame-level representations. Their output is consequently difficult for a viewer to verify and steer toward individual preferences. We present the semantic action graph, a lightweight domain schema that represents a sports match as performer, action, recipient, moment, and state nodes connected by role, temporal, and outcome edges. The schema demonstrates three key properties: 1) connected event sequences, 2) a shared, closed vocabulary, and 3) frame-addressable moments, making it suitable to serve two consumers at once: an agentic pipeline that composes narrated highlights, and a visual interface through which viewers query and inspect the same structure. We instantiate it in SportSAGE, a design probe pairing a four-module highlight pipeline with a graph interface, and report feedback from 12 soccer fans. Participants were satisfied with the quality of the generated highlights and narratives, and used the graph interface to search, navigate, and interpret the match highlights. These results provide early evidence that one small, human-readable schema can ground agent generation and support human interpretation at the same time.

World Action Models (WAMs) advance beyond conventional visuomotor policies by jointly predicting future world states and robot actions, enabling the policy to learn physical dynamics that support effective control. However, recent tactile WAMs often rely on large-scale pretrained generative backbones to capture contact-rich physical dynamics, which limit their inference efficiency and flexible deployment. In this paper, we present \ABBR{}, an agile tactile World Action Model for contact-rich robot control. \ABBR{} encodes visual and tactile observations into a shared latent that serves as the source of a direct vision-tactile-to-action flow-matching process, which can jointly generate latent representations of action chunks and future visual/tactile latents. A key observation is that vision and tactile signals evolve at inherently different timescales: adjacent visual frames are often highly similar, whereas tactile signals can change abruptly upon contact. We therefore introduce multi-horizon multimodal prediction in \ABBR{}, which provides supervision for visual latent at a larger temporal offset while predicting the tactile latent in the next frame to capture fine-grained contact dynamics. Across nine simulated and five real-world contact-rich manipulation tasks, \ABBR{} demonstrates strong and robust performance, outperforming the strongest baseline in success rate while maintaining low inference latency. In particular, in five real-world experiments, \ABBR{} yields a relative gain of $\textbf{29.4\%}$ in overall success rates while achieving inference latency of $\textbf{11.9 ms}$. These results demonstrate that multimodal WAM can be achieved with an agile architecture suitable for precise and high-frequency robot control. More details are available on our project page: https://hanchuzhou.github.io/TARO_project_page/.

Evaluating an AI system requires disaggregated assessment, as performance varies across domains such as benchmark task types or conversation types in deployed agents. Exhaustive testing is expensive, so evaluation rests on a sample of labeled units. We treat the evaluation set as a finite population and seek accurate point and interval estimates of each domain mean. Direct estimators, including prediction-powered inference (PPI), use only a domain's own labels and are imprecise where labels are few. Small area estimation addresses this problem, and we build on it to develop an integrated workflow for estimation and validation. For estimation, we propose prediction-powered smoothing (PP-S), a Bayesian model fit to each domain's prediction-powered estimate, with an extension that borrows strength across a reporting taxonomy (PP-TS). For validation, we derive a new, approximately unbiased design-based cross-validation score for choosing among direct and smoothed estimators. We study a curated benchmark with verifiable grading and deployed agent traffic graded by humans, each with every outcome observed. In both, the proposed estimators improve on the direct estimators in point and interval estimation, with near-nominal coverage. At the same sampling budget, our score selects as well as an independent validation sample does and estimates the selected estimator's error far more accurately.

As scaling pre-training data alone yields diminishing returns, post-training is becoming increasingly important across physical AI domains such as autonomous driving. End-to-end driving policies are pre-trained in open loop with behavior cloning on human demonstrations. However, compounding errors during closed-loop deployment can take the vehicle outside the training data distribution, increasing the risk of safety-critical incidents. Closed-loop post-training can mitigate this risk but requires costly simulation for sensor-based policies. We propose OPTED (on-policy fine-tuning for end-to-end driving) which decouples reinforcement learning from the post-training of the end-to-end policy: a privileged teacher is trained using RL on vectorized inputs (HD-map and bounding boxes). This teacher then provides supervision to the pre-trained student during closed-loop post-training. We apply OPTED to two camera-based models, TransFuser and VaVAM, and fine-tune them in AlpaSim, using neural reconstructions (3DGS) of real driving logs. Driving scores increase by factors of 1.6$\times$ and 9.5$\times$, respectively. In controlled experiments OPTED matches closed-loop performance with approximately three orders of magnitude fewer simulator interactions than direct RL post-training, while staying closer to the human prior. Project page: https://01dami23.github.io/opted/

Effective troubleshooting agents in enterprise customer support depend on retrieving actionable guidance from similar historical cases, yet existing retrieval-augmented generation (RAG) systems treat support cases as static documents and overlook their multi-stage, stateful nature. We introduce RAFT (Retrieval-Augmented Framework for Troubleshooting Agents), a stateful RAG framework that abstracts each closed historical case into a directed chain of timeline entries and retrieves at the entry level, surfacing cases whose intermediate states match the active case and returning the parent-case trajectory anchored at the matched state; an optional case-level graph links cases through a configurable similarity representation. We evaluate this retrieval layer directly, which, unlike evaluating a full agent system, requires no production deployment. Because public multi-stage troubleshooting data is extremely rare, we pair a synthetic benchmark built from Microsoft Learn Windows Server documentation with real Apache Jira issues carrying human-created duplicate labels. RAFT improves Case Hit over vanilla RAG and GraphRAG baselines at every stage of case progress, with statistically significant gains over the strongest baseline; the Jira results provide directional evidence that the advantage transfers to real case histories. We release our benchmark, implementation, and the Apache Jira evaluation set.

Falsification searches for counterexamples to formal specifications in cyber-physical systems (CPS). With specifications written in Signal Temporal Logic (STL), falsification can be formulated as a robustness optimization problem, traditionally tackled with black-box search algorithms. In parallel, large language models (LLMs) have recently emerged as surprisingly effective optimizers when coupled with iterative prompting. In this work, we connect these ideas and introduce LLM-Falsifier, an LLM-based approach that falsifies specifications by minimizing the STL robustness degree. Beyond generic prompt-based optimization, our key idea is to expose the LLM to semantic information that is natural for language models but absent from standard numerical optimizers, including natural-language input and output names, output trajectories, and critical-time witnesses for the minimum robustness value. These additions enable smarter and more sample-efficient robustness search. On the ARCH-COMP falsification benchmarks, LLM-Falsifier is shown to outperform existing falsification tools based on a range of optimization paradigms, from surrogate-based and Bayesian optimization to search-based testing, on 14 of 21 specifications when measured by the average number of simulations required to find a counterexample.

Reinforcement learning constitutes a promising approach owing to its potential for superhuman performance and self-learned policies. However, its application to real-world autonomous driving remains scarce, particularly in unstructured environments, because of the challenges associated with sim-to-real transfer for unstructured environments. In this work, we present MILER, an end-to-end policy framework with zero-shot sim-to-real transfer. During offline training, we employ a custom semantic mid-level representation (MLR) simulator and train the policy network using reinforcement learning, with its control outputs applied directly to a bicycle model. During deployment on the real vehicle, camera and LiDAR data are processed by BEVFusion to generate a semantic bird's-eye-view representation consistent with that of the MLR simulator. The actions generated by the policy network are not applied directly to the real vehicle. Instead, we employ a trajectory-alignment strategy that enables zero-shot sim-to-real transfer of both perception and control. We extensively evaluate the proposed framework on a diverse test track comprising numerous challenges, including various obstacles, hairpin curves, velocities of up to 33.6 km/h, and off-road sections. In total, we drove 17.3 km with two different vehicles on a 3.0 km test track without human intervention, thereby demonstrating the effectiveness of our approach. Furthermore, the entire software stack runs on a Jetson AGX Orin.

Video diffusion models repeatedly process long spatiotemporal token sequences during denoising, making attention a major computational bottleneck. Linear attention offers an appealing alternative and has been widely adopted in recent large language models, but directly applying it to video models often fails to preserve the fine-grained interactions required for high-quality generation. We present Video DeltaNet (VDN), which combines local Softmax attention with bidirectional linear memory for long-range video context. Its linear branch introduces Video Delta Attention (VDA), which updates memory once per frame by jointly incorporating its spatial tokens. Separate output projections and learnable gates calibrate the two branches, while a staged teacher-alignment recipe progressively introduces the new pathway into pretrained models. We instantiate VDN on MiniMax H3, applying the hybrid to video-to-video interactions while retaining Softmax for interactions involving text or audio. With eight-step distillation and an optimized SGLang serving stack, VDN-H3 completes DiT denoising for a 14.3-second, 768p video in 6.70 seconds on eight NVIDIA B200 GPUs, corresponding to a 14.5x speedup over the 50-step dense H3 baseline on the same GPU count.

Reasoning and agentic workloads increasingly demand efficient long-context inference. Yet full-attention decoding reads the growing history at every step, regardless of its benefit to the next prediction. We show that a pretrained model's decoding states already contain information predictive of this benefit, before the global read. Building on this finding, we introduce On-Demand Attention (ODA), a local-first decoding method that uses a lightweight recall head to selectively invoke global attention as its predicted benefit changes during generation. ODA trains only the recall head, leaving pretrained weights unchanged and the complete historical KV cache available for future recall. We further implement GPU-side conditional execution in vLLM, translating reduced global reads into practical decoding speedups over full attention at long context lengths. Experiments across Qwen and Gemma models, including hybrid-attention backbones, show that selective recall recovers most of the performance lost under local attention while substantially reducing global reads. These findings support long-context inference in which pretrained models guide their own access to the information they retain.

Semantic cell annotation improves chunking interpretability for spreadsheets in LLM-driven RAG systems, aiding answer generation through enriched context rather than improved retrieval accuracy. We propose a novel framework of splitting any spreadsheet into interpretable chunks using cell role annotation. Our framework beats the state of the art, yet it faces a hard ceiling. Spreadsheets are fundamentally two-dimensional unstructured data with continuous relationships and infinite potential cell roles. Because classification models are restricted to finite, pre-defined classes, they cannot perfectly capture this structural nuance, even with human-level annotation. We show that addressing the spreadsheet-to-LLM bottleneck requires moving beyond discrete cell classification. Instead, the field must develop dimensionality-reduction techniques to directly flatten 2D unstructured spreadsheets into 1D unstructured text. Text chunks would be easier for downstream RAG to interpret and generate from.

Activation steering modifies LLM behavior at inference time, but identifying where and how strongly to steer remains manual. We introduce Deep Noir, a framework that uses Logit Lens convergence and causal head-level attribution to autonomously discover optimal steering parameters. Across three scales (1B x 3, 2-3B x 2, and 7-9B x 4), our engine achieves 16.7 percentage-point improvement on spam at 1B (standard deviation 4.7; 39 runs), with gains increasing to 21 to 42 percentage points at 7-9B across four architectures. On SST-2 sentiment, it achieves a 13.1 percentage-point improvement with zero code changes. Mechanistic grounding enables automated discovery of intervention points that generalize across tasks and architectures. On sentiment, RepE without head masking fails to improve over baseline, while Deep Noir improves all models (p less than 0.01). We further show that steering creates a predictable prompt-injection attack surface whose vulnerability increases monotonically with steering magnitude. This finding is relevant to agent systems deploying steered classifiers.

Agent trajectories record what an agent does and what happens next. Yet standard supervised fine-tuning (SFT) applies loss only to agent-authored action tokens, using environment observations as context but not as prediction targets. We ask whether this convention provides the best initialization for subsequent reinforcement learning. We introduce ActObs, which also supervises the observation tokens already present in each trajectory. Although deployed agents never generate observations, learning to predict them encourages the policy to model action consequences without adding data, parameters, sequence tokens, or forward passes. The methods perform similarly after SFT but diverge after GRPO. On Qwen3-4B, GRPO from ActObs achieves higher pass@k at every evaluated sampling budget than its action-only counterpart on Terminal-Bench 2.0. On Qwen3-8B, it trades some pass@1 reliability for higher pass@k (+3.4 pp at pass@16) and solves more distinct tasks. The advantage extends to cross-domain code editing on aider-polyglot (+4.2 pp at pass@1 at 4B), whose tasks are unseen during SFT and RL. ActObs retains more entropy during RL while requiring less policy movement, leaving the final policy closer to its SFT initialization. Our analysis traces this difference to SFT: action and observation gradients rapidly become orthogonal, while action-only training leaves a large residual observation gradient and degrades environment prediction below the base model. Joint supervision prevents this one-sided specialization, preserving consequence prediction and preparing the policy for downstream exploration.

This paper introduces and operationalizes summarization bias: a proposed systematic tendency of large language models (LLMs) to represent narrative meaning as an abstract summary label rather than as the reconstructable inferential structure that produces it. Within the Bulut Doctrine, narrative effect is theorized along a told-shown axis: in told mode, emotional and informational content is declared explicitly and requires little reader reconstruction; in shown mode, that content is suppressed at the surface and must be reconstructed from physical cues and indirection (Objective Projection). Shown mode is the higher-load condition the doctrine is designed to measure. The claim is that LLMs fail along this axis in a specific direction. Summarization bias is hypothesized to operate in two regimes: (i) a generative regime, in which a model asked to render an emotion through Objective Projection defaults to declaring it instead; and (ii) an evaluative regime, in which a model judging narrative quality rewards told-mode explicitness and under-detects shown-mode suppression. The evaluative regime is the more consequential, since LLMs increasingly serve as judges and reward models, and a directional bias toward told mode would impose a selection pressure degrading prose toward flat declaration. This report does not claim the bias is validated. It defines the construct, situates it against LLM-as-judge biases, rereads a completed independent reliability study as directional evidence consistent with it, and pre-registers a two-regime test with decision rules under which the construct would be abandoned.

World Action Models (WAMs) improve robot policy learning by incorporating future dynamics, yet explicitly generating future videos at inference introduces substantial computational overhead. Removing future generation improves efficiency, but leaves future dynamics only implicitly encoded in observation features, which can limit robustness under distribution shifts. We propose MoWAM, an efficient WAM that replaces future video generation with explicit future motion prediction. Instead of reconstructing the complete future scene, MoWAM models structured robot motion as a compact abstraction of the future, capturing how the robot is expected to evolve under the current scene and interaction constraints. A Mixture-of-Transformer architecture learns future visual dynamics during training while jointly predicting motion and action, allowing video generation to be removed entirely at inference while retaining an explicit representation of the future. The compact motion representation further enables efficient inference-time scaling by sampling multiple candidates of motion and action pairs and selecting among them with a motion-aware task-progress verifier. Experiments on LIBERO, LIBERO-Plus, and real-world manipulation tasks demonstrate that MoWAM achieves strong in-distribution performance, improved out-of-distribution robustness, and higher average real-world success than representative WAM baselines. In addition, performance improves as more candidates are explored, demonstrating that explicit future motion provides an effective and efficient basis for inference-time scaling.

Episodic test-time adaptation resets a frozen segmenter to source weights $M_0$ on each case and adapts for a fixed step count. A fixed horizon conflates a cohort-level question, how far to adapt, with an irreducibly per-case one, whether this case should be adapted at all. Cohort means hide that decision: on cross-vendor cardiac MRI the mean $Δ$Dice from adaptation is statistically indistinguishable from zero while 58.7% of cases are individually made worse. We quantify this harm as harmful accepted area (HA), the harmful fraction of the edited area a controller deploys. Held-out tuning gives a stronger baseline than a fixed horizon, but the budget it selects transfers on neither of the two main medical benchmarks, and no global budget can condition on the case. We show that prediction fragmentation---the disagreement geometry between $M_0$ and the adapted mask $M_k$---predicts HA with no labels or extra backward passes at decision time, comparably on three benchmarks (Spearman $ρ$ 0.50--0.60), at a quarter of gradient-norm's latency. A case-level router built on it cuts HA from 0.228 to 0.139 on a benchmark that took no part in its design, with the design frozen and only cut-points recalibrated there. On the cardiac benchmark the design was selected on, the router cuts HA from 0.129 to 0.013 at matched Dice and 1.10 deployed updates, against the retrospective-best budget found post hoc on evaluation labels, and reduces that 58.7% to 20.0%, an upper bound we quantify. Where the retained cases are not net-helped (as on prostate), the router still cuts HA but concedes accuracy, a boundary we report. Thresholds are fit once on a labeled split disjoint from evaluation; decisions use no labels or gradients. The template ports across architecture and domain (nnU-Net$\to$SegFormer, Cityscapes$\to$ACDC) with coordinate, thresholds and per-bucket actions instantiated per domain.

Mission-level autonomy for disaster response, search and rescue, and tactical UGV operations requires repeated path and mission planning as terrain conditions, agent types, and objectives change. Pixel-grid search is costly for repeated kilometer-scale queries, while semantic abstractions must maintain valid costs and connectivity as conditions change. We present HOPHY (Hierarchical Off-Road Planning using Hypergraphs), a reusable hierarchical terrain representation that organizes map-scale terrain into geometrically connected semantic regions (GSNodes), connectivity-preserving critical regions (Coarse Regions), and typed hyperedges for terrain, agent, and weather context. Hyperedge intersections select affected regions and incident edges for state updates without rebuilding the hierarchy. Across real off-road maps spanning kilometer-scale areas, HOPHY achieves 100% planning success and less than 0.01% median cost deviation from the oracle (pixel A*), with substantially lower query and replanning latency than the evaluated pixel and abstraction baselines. Applied to a multi-robot task-allocation (MRTA) problem, these gains reduce total computation by 79x over pixel A* and 7.2x over the fastest abstraction baseline, with mission makespan comparable to pixel A*. Finally, we demonstrate HOPHY on a physical Clearpath Jackal that successfully executes a 1.5-km, eight-task mission across mixed-surface outdoor terrain and a blockage-triggered replanned route.

We study first-order black-box convex optimization over an $\ell_p$-ball for objectives Lipschitz in the $\ell_q$-norm, solving in the affirmative the nonsmooth version of the COLT open question (Guz15b) on whether the geometry of a smaller feasible set ($p < q$) can improve convergence rates in convex optimization, and matching prior lower bounds up to logarithmic factors. Our rates include \(\widetilde O(1/T)\) for convex Euclidean-Lipschitz optimization over the $\ell_1$-ball, improving on the $O(1/\sqrt{T})$ classical rate under general assumptions. The key technical device is a new online learning game, where the comparator is evaluated using the maximum of affine losses observed so far. We bound the value of this game above and below in terms of a combinatorial online learning quantity: the sequential fat-shattering dimension, which we characterize for the $\ell_p / \ell_q$ case. Our results generally apply when the feasible set $X$ and the set of possible subgradients $H$ are convex, centrally symmetric, and admit a type of minmax theorem, advancing on a fundamental question by Sridharan [Sri12, Section 10.1.2, Q3]. As a geometric consequence of our analysis, of independent interest, we obtain estimates for the expected distance of a convex hull of samples to their mean in several Banach geometries, a version of the celebrated Wendel's theorem (Wen62), but quantitative and for bounded general distributions as opposed to centrally symmetric ones.

Scaling optical input/output for co-packaged optics is limited by the fiber-to-chip interface. Conventional edge coupling offers low loss and broad bandwidth but confines channels to a single row along the chip facet. On the other hand, surface couplers are either narrowband or difficult to fabricate with high yield and, importantly, compete with back-end metal routing. In this work, we introduce volumetric edge coupling, in which the chip edge is structured in three dimensions so that optical coupling can occur over a two-dimensional region rather than along a single line, providing a route toward interfacing multiple waveguides with multiple cores of a multicore fiber while preserving optical access from the chip perimeter. We investigate a single-channel realization based on total-internal-reflection (TIR)-mediated evanescent coupling through the 54.7$^\circ$ sidewall of a KOH-etched silicon cavity, with the coupling profile shaped by a wedged buried-oxide ridge. Finite-difference time-domain simulations predict a peak coupling efficiency of 88% (-0.56 dB) at 1550 nm and a 1-dB bandwidth of 86.45 nm spanning the C-band. The design further exhibits a vertical alignment tolerance of approximately $\pm$2 $μ$m, weak sensitivity to transverse offsets up to 5 $μ$m, and a $\pm$0.8$^\circ$ 1-dB angular tolerance. The reflected optical field also provides a potential alignment signal, offering a path toward reduced active alignment overhead in future multicore implementations.

Large language models are increasingly used in healthcare communication, yet most evaluations emphasize response quality while assuming that the user's concern has been interpreted correctly. We introduce HerHealthEval, a controlled evaluation framework for multilingual understanding of women's-health communication. For each clinical case, HerHealthEval provides matched versions in English, French, and Modern Standard Arabic using six communicative forms: canonical, clinical, layperson, indirect or hedged, emotionally concerned, and deliberately under-specified. The first five express the same underlying concern and retain the same clinical information, whereas the under-specified form intentionally omits relevant details to test whether the model recognizes that clarification is needed. We evaluate a multilingual instruction model and QLoRA-adapted variants on concern classification, risk calibration, clarification behavior, parse compliance, and cross-form consistency. Results reveal that aggregate accuracy and consistency can conceal safety-relevant failures. A multilingual adaptation model reaches 0.994 under-triage in French and Arabic under language-asymmetric risk supervision. A controlled re-adaptation using source-derived, language-invariant risk labels reduces under-triage to 0.572 and 0.558, respectively. These findings show that robust multilingual healthcare evaluation requires explicit testing of register variation, uncertainty handling, and the provenance and invariance of adaptation labels.

Scalable machine learning in challenging underwater environments is strongly limited by the lack of labeled real-world training data. This data is often expensive and laborious to gather, making large-scale real-world data challenging to gather and curate. However, simulated data can help close the gap, enabling many learning-based tasks for underwater perception. In this work, we extend OceanSim, an IsaacSim-based underwater perception simulator, with a Synthetic Data Generation (SDG) pipeline for training models to be used in underwater scenarios. The proposed pipeline enables users to generate large, automatically labeled, photorealistic datasets with configurable scene appearance, structure, and sensor settings. We evaluate the pipeline on a real-world sea urchin detection task and study how different forms of synthetic scene variation affect sim-to-real performance. Based on these experiments, we discuss findings on our results, main limitations of the current pipeline and identify future directions for improving underwater rendering fidelity, scene diversity, and the evaluation of sim-to-real generalization. The open-source code can be found at https://github.com/umfieldrobotics/OceanSim.

To operate effectively in human environments, robots must identify articulated objects, segment their movable and interactive parts, and estimate their kinematic models. Existing articulated scene representations typically recover kinematics from observed interactions, while methods operating on static scans often decouple articulation from functional interactive elements. We present FunArt, a framework that constructs articulation-aware functional 3D scene graphs from posed RGB-D observations captured in a single static configuration. FunArt reconstructs object instances, converts their fused geometry directly into the O-Voxel representation of TRELLIS.2, and exploits its frozen, sparse-compression VAE as a structural prior. A lightweight query-based decoder combines compact object-level latents with dense, surface-aligned features to jointly segment movable parts and functional interactive elements while estimating motion type, axis, origin, and range. On the Articulate3D dataset, FunArt achieves state-of-the-art performance across movable-part segmentation, articulation estimation, and functional-element segmentation, both with and without ground-truth object input. In the end-to-end setting, it outperforms the strongest baselines by 1.5 AP_{50} points for movable parts, 2.8 AP_{50} points under joint origin-and-axis constraints, and 6.7 AP_{50} points for functional elements. These results demonstrate that generative 3D latents encode actionable structural cues that can initialize robotic perception and planning before physical interaction.

3D diffusion policies are strong at generating geometrically grounded actions from current observations, but successful manipulation requires not only knowing what motion is feasible now, but also anticipating where the interaction is heading. Existing policies largely leave such foresight to emerge implicitly from action learning. We introduce Movement Trend Guidance, a simple but effective way to provide this foresight without introducing an explicit plan. From a short observation history, the policy learns a compact latent representation of interaction evolution. During training, sparse future gripper states supervise this representation; at inference, only the latent is retained as future-oriented conditioning alongside the current observation. The latent provides global conditioning for action generation, while an additional gated FiLM branch is used only at the UNet bottleneck. Despite adding only 3.52% more parameters to DP3, our method preserves the original dense-action and receding-horizon formulation and consistently improves upon DP3 across RoboTwin2.0, LIBERO-40, and DexArt. It reaches 62.8% vs. 56.1% in 50-task RoboTwin2.0 mixed training, 71.93% vs. 37.08% on LIBERO-40, and 72.0% vs. 49.0% on five real-robot tasks. These results show that a diffusion policy can benefit substantially from knowing where an interaction is heading, without being told exactly where to move.

Large-scale vision-language-action (VLA) models provide powerful priors for robot manipulation, yet adapting them to a specific deployment remains challenging. Supervised fine-tuning (SFT) on task-specific demonstrations provides a step toward deployment, but faces two persistent limitations: static data provide limited coverage of out-of-distribution states, and standard imitation objectives do not distinguish progressing behavior from less useful data. Interactive post-training can address these limitations, but typically requires repeated policy execution and human intervention on a physical robot. We introduce HIL-UMI, a policy-guided Universal Manipulation Interface (UMI) framework for robot-free human-in-the-loop VLA post-training. During handheld UMI demonstrations, HIL-UMI queries the current policy on the same observation stream without executing its predictions. The Energy Score compares the human action trajectory with policy inference and triggers collection when their discrepancy indicates an out-of-distribution region. In a separate feedback loop, low online advantage predictions identify essential segments for refining a progress-based advantage estimator. The updated estimator then guides advantage-conditioned behavioral cloning using a balanced mixture of base demonstrations and new policy data. This design preserves the iterative and policy-aware nature of human-in-the-loop learning while decoupling data collection from robot deployment. Experiments on four real-world tasks spanning long-horizon and precise manipulation show that HIL-UMI achieves consistent improvement over SFT and benefits from both targeted collection and advantage refinement. Moreover, HIL-UMI outperforms HG-DAgger on Clean Up Table with lower per-frame collection time, suggesting a scalable path for VLA post-training across operators and locations.

Learning predictive models of contact-rich dexterous manipulation requires dense tactile interaction, but such data are costly to scale on real robots and remain tied to embodiment-specific sensors. We introduce DexTouch-WM, an action-conditioned world model that learns from scalable human touch to jointly predict future RGB observations and bilateral tactile dynamics. Our insight is that human and robot manipulation share transferable contact dynamics when their tactile observations and action spaces are made compatible. We deploy flexible piezoresistive arrays with a shared sensing layout on both human and dexterous robot hands, and retarget human motion into the robot action space so that human interaction can supervise the same dynamics model used for real-robot prediction. DexTouch-WM couples a pretrained video expert with a lightweight tactile expert using anatomy-aware tactile tokens and aligned action conditioning. In human-to-robot scaling experiments, we keep five hours of real-robot supervision fixed while increasing human interaction from 0 to 100 hours, and observe substantial improvements in held-out robot-domain visual, geometric, and contact prediction despite disjoint human and robot task sets. Beyond prediction, we evaluate the world models as surrogate environments for policy evaluation and as generators of synthetic trajectories for real-robot policy learning, showing that scalable human interaction provides a complementary data axis for learning dexterous robot world models.

Vision-Language-Action (VLA) models are a class of generalist robot policies that map camera images and language instructions directly to robot actions. While promising, these models remain slow at test time, particularly for long-horizon tasks that require many queries to the policy. Recent efforts reduce VLA latency by distilling smaller models, overlapping asynchronous action chunks, or pairing the VLA with a fast low-level policy, but still run a learned policy for the entire task. In contrast to VLA, classical motion planners quickly find collision-free motions, but require an explicit goal and have no semantic understanding of the task. In this work, we present SkipVLA, a hybrid policy that combines a pretrained VLA with a classical motion planner, using the planner for free-space motion and querying the VLA only for contact-rich skills such as grasping and placing. SkipVLA reuses the frozen vision-language backbone of the VLA to predict a target pose for each planned motion, and learns this predictor without additional demonstrations introduced into the system by using what was already learnt by the large VLA. We evaluate SkipVLA with three VLAs on 13 LIBERO tasks in simulation and three pick-and-place tasks on a physical 6-DoF YAM arm, demonstrating up to 2.5x faster task completion and significantly lower energy consumption while achieving the same task success rate.

A vision-language-action (VLA) policy with a flow-matching action expert generates each action chunk (a short command sequence) by integrating a learned velocity field; once its weights are fixed, the success or failure of an earlier rollout cannot change the chunk generated now. Concurrent test-time methods give a frozen policy such an input from retrieved successes, a learned critic, a verifier, or a dynamics model, but none uses the robot's own failed rollouts as negative evidence with nothing but a terminal outcome bit. We introduce TraceFlow, a progress-aligned guidance field that turns the action densities of retrieved successful and failed rollouts into a bounded correction to a frozen flow-matching action expert, using one terminal outcome bit per rollout and no other label. Its TraceBank stores traces, time-ordered state-action records with a terminal label, starts from the target-task training traces, and later admits the deployed robot's own rollouts. On an ordered real-robot packing task the base completes 21 of 50 trials in order, TraceFlow 39, and one stacking round without any weight update 47, with wrong-sequence episodes falling from 20 to 0. In simulation the gain is selective: with per-suite selected settings, TraceFlow raises RoboMemArena Sequence from 78.92\% to 91.50\% task success and Transferring from 54.41\% to 62.00\% at stacking round 2, leaves the 26-task aggregate unchanged, lowers Counting and Occlusion by 1.12 and 1.42 points, and changes LIBERO-Plus (Long) by +1.27 points (p = 0.0733). Stacking gains are finite, every branch peaking before round ten, and the bank's success-to-failure ratio predicts no retrieval allocation.

Large language model responses are non-deterministic, so failures in LLM agents are hard to reproduce: a failure depends on inference that is not bitwise reproducible, on tools that read changing state, and on a multi-step trajectory that a re-run rarely repeats. Record-and-replay makes a run reproducible, but existing agent tooling records runs only to trace or score them, not to test a code change against them. We present Chronicle, which records an agent run at its non-deterministic boundaries as immutable envelopes and replays it from the record. Its central operation, cut-point replay, serves a chosen subset of boundaries from the record and executes the complementary subset live with new code, turning a recorded incident into a regression test that runs in continuous integration. On a benchmark of 6 recorded failures with simulated model boundaries, recording adds 23 μs per crossing (0.008% of an assumed 300 ms model call), full replay issues zero model calls and is bit-stable across 20 repetitions, and cut-point tests fail on faulty code and pass on guarded and benign changes for all 6 incidents. In a mutation study of the guarded tools, cut-point tests catch every mutant that lets the recorded unsafe action through, while a baseline that stubs every boundary, using the same assertion, catches none. Chronicle and the benchmark are publicly available at https://github.com/theagentplane/chronicle.

A robot sent to a named gas leak must preserve gas identity, estimate the source, and navigate to the resulting goal. We present SmellDiffusion, a simulation pipeline that represents species-specific gas zones in an open-vocabulary olfactory scene graph and shares the selected goal between classical and diffusion planners. Its key components are a peak-local geometric gate for selective source correction and diffusion-based, gas-guided trajectory generation. Among 424 unique source-wind configurations in solved flow, 28 have a concentration peak displaced more than 0.5m from the source. A source-independent geometric gate, calibrated only on the training split and evaluated at the observed peak, detects 9 of 10 held-out displacements at 0.64 precision. Gating a precomputed forward-matching correction reduces mean error on the displaced cases from 1.468m to 0.592m (60%), using matching for only 14/204 cases. All-case mean error falls from 0.205m to 0.180m. All planners receive the same scene-graph source estimate as their goal. In a controlled comparison, best-of-ten diffusion achieves mean gas exposure comparable to gas-guided A* (0.0476 versus 0.0455). A single diffusion proposal takes 41.7ms, compared with 72.3ms for gas-guided A*, although best-of-ten sequential sampling increases total runtime. Plain A* also reaches the same goal and remains the fastest and shortest-path method. Six matched Gazebo runs give mean robot-to-source errors of 0.39m for A* and 0.31m for diffusion.

Recent single-view feed-forward 3D Gaussian Splatting (3DGS) generation predicts a fixed number of Gaussians per camera ray, introducing severe spatial redundancy. Most existing compaction strategies target multi-view setups to exploit cross-view consistency and are incompatible with single-image models. Instead of retraining the base feed-forward network to directly output compact representations, our insight is to keep the base models frozen and apply post-hoc pruning and recurrent refinement to the generated Gaussians. Consequently, we propose a backbone-agnostic compaction pipeline for single-view feed-forward 3DGS that couples an importance-score-based pruning mechanism with a trainable, lightweight recurrent refinement module, which iteratively updates the surviving primitives to restore image quality. Our results demonstrate seamless integration with existing baselines while preserving novel-view rendering fidelity and achieving high memory reduction. Furthermore, our method supports flexible inference-time keep ratios for application needs.

We propose a novel surrogate model for linear coupling in graded-index multimode fibers that is computationally more efficient than existing models, and can be tuned through a single measurable parameter. The model is benchmarked numerically and experimentally on a 90-polarization-mode fiber.

Frontier AI models are rapidly gaining the ability to exploit vulnerabilities in complex pieces of software. The risk is not theoretical, as evidenced by recent sandbox escapes performed by frontier models at OpenAI and Anthropic. Discussions of how to sandbox inference stack components often focus on components other than the inference engine itself (e.g., network proxies or code execution environments). However, the inference engine is an attractive target for a misaligned model. For example, if a model can trigger exploits in that engine merely by generating specially-crafted output tokens, the model can initiate a multi-step, to-the-bare-metal exploit chain in the engine, without relying on vulnerabilities in other components of the inference stack, and without assistance from externally-provided, maliciously-crafted input tokens. In this paper, we show that a misaligned model can perform inference engine fingerprinting to determine the specific engine (e.g., vLLM, SGLang) which executes the model. Once the engine has been fingerprinted, the model can leverage engine-specific exploits to take control of the engine using only carefully-selected output tokens. We provide concrete examples of model fingerprints in five popular engines, and demonstrate how realistic agentic harnesses allow a model to leverage those fingerprints to identify the local engine. We also describe a proof-of-concept, to-the-bare-metal exploit chain that originates from a fingerprinted (and subsequently compromised) inference engine. We conclude by discussing several ways that inference engines could be changed to make fingerprinting attacks more difficult.

On-policy self-distillation (OPSD) lets a language model learn from a frozen copy of itself that sees an answer or a worked solution. Giving the teacher this extra information seems to offer the student more to learn, but how much does it add beyond distillation itself? To isolate that contribution, we construct AMPLE-Math, a reusable suite of 5,319 mathematical problems with six reasoning views that share the same answer, and compare each view with matched reference-free distillation. With a thinking-enabled teacher supervising direct-response rollouts, reference-free distillation accounts for much of Qwen3-1.7B's improvement under thinking-enabled evaluation, both in domain and on external benchmarks. Evidence for an additional reference benefit is modest in Qwen, strongest for a polished solution, whereas complete traces add two percentage points in SmolLM3-3B at step 50. These benefits depend on the student being trained. At the same checkpoint, replacing short direct-response rollouts with long thinking-enabled rollouts turns gains into losses in both families while the problems, references, and evaluation stay fixed. Teacher profiles and matched loss interventions in Qwen further show that changing token-level supervision can leave student behavior largely unchanged. Together, these findings suggest that OPSD can improve access to existing reasoning capabilities through parameters shared by direct-response and thinking-enabled inference. The value of a privileged reference is what it adds to this cross-mode transfer, not how much of the solution it reveals.

This paper presents a real-time semantic world modeling framework specialized for precision agriculture using autonomous robots. The framework combines probabilistic mapping of objects and their semantic attributes, updated through Bayesian inference, with a graph-based Simultaneous Localization and Mapping (SLAM) approach implemented using $g^2o$, a general framework for graph optimization. This integration enables accurate mapping and localization without relying solely on GPS. By leveraging semantic information such as plant type, size, and health, the robot can perform tasks while mapping and localizing itself within a field of crops. The proposed framework was validated through Gazebo simulations and physical experiments on an indoor field with artificial plants using Boston Dynamics' robot dog Spot. A YOLOv8n object detection model was trained to extract object and semantic data from depth camera observations. These simulations and experiments demonstrate that the system can successfully perform real-time mapping of up to at least 400 plants.

In this paper, we tackle the problem of jointly estimating the system states and partially unknown dynamics within distributed sensor-equipped networks, particularly in scenarios where only partial state observations are available. To address this issue, we propose an observer-based dynamic cooperative learning framework incorporating online distributed Gaussian Process (GP) regression, which enables accurate estimation despite incomplete in measurements and deficient GP models. In addition, a novel data collection strategy is introduced, with theoretical conditions ensuring feasible data acquisition. Moreover, we also derive an error upper bound encompassing state estimation and model estimation, leveraging the deterministic error bounds of GPs. Empirical simulations demonstrate the superiority of our approach compared to existing distributed GP-based methods.

Quantum long short-term memory (QLSTM) models extend recurrent sequence learning with variational quantum circuits, but their optimization behavior can vary substantially across random initializations and temporal contexts. This paper evaluates a recursive QLSTM architecture against a standard QLSTM for one-step-ahead prediction of daily minimum and maximum temperature. Using daily weather observations from Toronto and identical training settings, we compare convergence, predictive accuracy, and generalization across input windows of 8, 16, and 32 days over 20 random seeds. The recursive model consistently reaches a near-optimal test loss earlier, reduces mean absolute error and root mean squared error, and exhibits a smaller generalization gap. These results indicate that recursive quantum feature transformations can improve stability and out-of-sample performance for compact hybrid quantum--classical temporal models.

Referring scene understanding for embodied robots requires grounding object- and relation-centric language queries from a designated viewpoint. While a local semantic Gaussian map can support such grounding within one agent's observations, cooperative settings require this ability to remain effective after independently reconstructed maps are aligned and fused. In this setting, the referred target or its contextual landmark may come from another agent's observations, while spatial relations must still be interpreted from the querying robot's viewpoint. We formulate this problem as cooperative referring Gaussian grounding over fused maps, which requires geometric alignability, instance-level semantic comparability, and view-conditioned relation reasoning. Existing language-aware Gaussian methods mainly focus on single-map querying, whereas Gaussian registration methods optimize geometric or photometric alignment without preserving language-grounding-oriented semantic compatibility. We propose CoRef-GS, a cooperative referring Gaussian splatting framework. CoRef-GS constructs local open-vocabulary instance-aware Gaussian maps, then aligns partially overlapping maps with a cross-agent alignment module by geometric and semantic consistency, and grounds queries using a view-conditioned mask relation graph. We further introduce CoQuad-Ref, a dual-quadruped benchmark spanning both real-world and simulated indoor scenes. Experiments show that, on simulated scenes, CoRef-GS reduces the rotation error from 2.58° after coarse initialization to 0.15° after refinement, and improves real-world referring mIoU over ReferSplat from 52.6% to 68.8%. The established benchmark and source code will be publicly released at https://github.com/ruojiruoli17/CoRef-GS.git.

Large language models (LLMs) are increasingly considered for safety-critical engineering, yet their reliability in regulated functional-safety workflows remains underexplored. We introduce SAFARI (Safety-Aware Functional Automotive Risk Inference), the first industrial benchmark for LLM-assisted automotive Hazard Analysis and Risk Assessment (HARA) under ISO 26262. It contains 3,000 de-identified industrial HARA cases and evaluates two coupled tasks: open-ended hazard analysis and standards-grounded risk assessment. To evaluate open-ended HARA artifacts, we propose the first reference-anchored LLM-as-a-judge protocol with high expert correlation. Experiments with nine frontier LLMs show that models often produce plausible hazard narratives but remain weak at ISO 26262 risk classification, with the best ASIL macro-F1 reaching only 0.261. Chain-of-Thought prompting provides limited benefit and often degrades categorical risk assessment. Error analysis further localizes major failures to scenario-critical context omissions during hazard generation and to controllability misjudgments during risk assessment, indicating where expert oversight should be concentrated. The dataset can be obtained from https://github.com/xixi47520-hash/HARA.

Human manipulation videos provide rich motion and interaction cues for acquiring robot skills without robot demonstrations. Video generation models synthesize such demonstrations from an initial scene image and task instruction, avoiding the need to record demonstrations for each task. However, the recovered motion captures only one scene-specific realization, leaving task structure, geometric relations, and constraints implicit. We present V2-STRep, a zero-shot framework that converts generated video motion into reusable robot skills through VLM-grounded structured task representations. The representation specifies motion phases, references, and task-relevant constraints, with targets described by minimal geometric structures: points, point-normals, axes, planes, and full 6D poses. VLM-provided 2D image-space cues are lifted into 3D using RGB-D observations to reconstruct task geometry and candidate grasp poses. Geometry-specific rules transfer motion to new scenes, while task-constrained trajectory optimization couples grasp selection with complete robot motion planning. It preserves task requirements while using remaining rotational freedom to accommodate joint limits. Updating deployment grounding and constraints enables reuse under new compatible instructions without generating another video. Experiments on six real-world manipulation tasks demonstrate improved execution success over baselines, reliable cross-scene transfer of successfully acquired skills, and adaptation to changed deployment instructions.

Learning visual policies for locomotion and manipulation requires coordinating contact with the environment and can incur substantial computation and GPU memory costs. First-order policy gradients (FoPG) reduce training cost through differentiable simulation, but local optimization can converge to unintended contact patterns. To address this shortfall, we propose Sampling-Guided Policy Search (SGPS), which couples recurring action-target refinement by sampling-based model-predictive control with first-order policy optimization. Behavior cloning initializes the policy from sampled actions; training then alternates sampling-based refinement with short-horizon FoPG updates under perturbed initial states and randomized dynamics. For visual policy training, we use a decoupled FoPG formulation that excludes rendering from the computation graph, enabling direct learning from depth observations without a state-policy teacher. On a single GPU, SGPS learns policies for locomotion, obstacle traversal, crate pushing, and bimanual carrying on simulated Unitree Go2 and G1 robots. Our experiments further show that refinement improves policy learning beyond initialization and tracking alone. For hardware deployment, the distilled policy transfers zero-shot to a real Go2 and uses onboard depth to autonomously trot, crawl, clear hurdles, and switch between these behaviors.

Answer grounding in document visual question answering remains an open challenge: most benchmarks lack grounding annotations or provide limited-quality labels, while constructing grounded datasets still requires costly manual effort. We introduce DocAttriBench (DAB), a large-scale benchmark for fine-grained, element-level source attribution in Document VQA, grounding answers to specific layout elements such as text blocks, tables, and images. To build DAB, we propose a Mask-based Perplexity-Derived Attribution method (MAPPET) that combines document layout and language modeling to identify the most informative element for each answer. MAPPET measures the increase in perplexity after masking candidate elements and attributes the answer to the element contributing most to model confidence. Applying MAPPET to multiple existing Document VQA datasets yields DAB, with 237k documents and 296k question-answer pairs with element-level grounding. We benchmark grounding-capable multimodal LLMs on DAB, evaluating answer accuracy, attribution accuracy, and overall answer quality. Results show that while larger models generally achieve higher answer accuracy, even the strongest models often fail to localize the supporting elements. DAB provides a scalable benchmark for developing grounded, verifiable, and trustworthy Document VQA models. Dataset and code are available at https://aimagelab.github.io/DocAttriBench/.

Animal demonstrations provide quadruped robots with natural and distinctive gait styles that are difficult to specify through hand-crafted rewards. However, their narrow directional coverage leaves little style-consistent supervision for backward, lateral, and turning commands. We present OmniMimic, a training framework that turns directionally limited animal demonstrations into a single multi-gait policy over target per-axis velocity ranges. OmniMimic first combines temporal reversal, constrained dynamics completion, and sagittal reflection to construct robot-specific kinematic and physical supervision beyond the observed directions. It then expands commands progressively from the demonstrated velocity distribution toward the target per-axis bounds, and uses a shared actor with soft-gated, gait-specialized residual experts to balance reusable locomotion skills with gait-specific corrections. Across four gaits in simulation, OmniMimic reduces mean foot-position RMSE at forward and backward reference velocities by 12.9% and velocity-tracking RMSE on a uniform Cartesian command grid by 63.1%, compared with the matched APEX baseline. The project page is at https://OmniMimic.github.io.

Roofing requires workers to coordinate locomotion, balance, and work-related body motions on pitched surfaces, creating a challenging application for humanoid robots. Directly retargeted human demonstrations, however, may preserve motion appearance while placing the robot's feet or hands incorrectly relative to the roof. This study presents a task-semantic scene-grounded framework for learning roofer-style whole-body motions on a Unitree G1. Human demonstrations are captured using a tracking system and retargeted to the robot, while a metric roof model supplies the spatial reference unavailable from the tracking system. A trajectory-level optimization grounds inferred support contacts and annotated work relations to the roof, and execution-aware reinforcement learning encourages the resulting policy to preserve these relations under dynamic tracking errors. The framework is evaluated through a multi-motion tracking study, a roof-pitch coverage matrix, a five-way nailgun ablation, cross-task experiments on hammering and lateral pushing, and comparisons with pure reinforcement learning and zero-shot teleoperation. Our method enables the robot to satisfy support, work-clearance, and nonpenetration criteria across all evaluated seeds. Across nailgun, hammering, and pushing, it achieves work-clearance errors between 0.256 and 0.531 cm and 3/3 successful evaluations per task. Physical experiments reproduce uphill walking, nailgun, hammering, and bending motions with mean base-frame motion errors below 80 mm. These findings establish scene-grounded human motion learning as a promising basis for construction-oriented humanoid motion primitives.

Full-consensus rates are often treated as indicators of collective cognition, yet depend on how participation and final states are operationalized. We replayed 100 held-out human Wason groups with matched large language model (LLM) agent groups, seeding one belief-anchored agent per participant's pre-discussion answer and scoring agents and people with the same code. Across human scoring definitions, estimates ranged from 24.0% to 57.0%; about one fifth of participants never posted, whereas agents almost always did. Agent groups remained more consensual in two post-unblinding sensitivity analyses: the submit-based comparison (n = 98) yielded gaps of 34.0 and 43.9 percentage points for chat and reasoning modes, and the participation-matched comparison (n = 45) yielded gaps of 34.1 and 44.4 points. These complementary routes reduced different measurement asymmetries yet converged within 0.5 percentage points. The gap persisted without early stopping and under a reparameterization removing the memorizable answer; reasoning-mode groups then agreed nearly unanimously, mostly on incorrect answers. Simulated consensus did not track collective accuracy, and belief-anchored agent groups were biased estimators of the human group-outcome distribution in this setting. These analyses provide a scoring-explicit basis for assessing simulated-group estimates of human deliberative outcomes.

Large language models can report a numerical confidence together with generated content, but it is unclear whether this report is more than calibrated rhetoric. We analyze three training-free signals: confidence verbalized with the answer, post-hoc $P(\mathrm{True})$, and agreement with three additional generations on the same 100 TriviaQA questions for two model families. Direct verbalization is a surprisingly strong baseline: after auditing benchmark errors, it reaches AUROC 0.956 and 0.937 for correctness prediction. Three-sample agreement is substantially weaker (0.765 and 0.790), and a fixed interpolation with verbalized confidence has no statistically reliable benefit. Four of nine errors from one model and two of eight from the other receive unanimous sample support, showing that self-consistency can amplify shared misconceptions. Re-eliciting confidence for the same fixed answers with equivalent prompts changes scores by 0.043 to 0.084 on average and flips 4\% to 9\% of decisions at a 0.8 threshold. An exploratory audit of 100 confidence-tagged biography claims further finds only a modest confidence gap between supported and contradicted claims. These results argue that useful self-reports remain sensitive to elicitation, correlated errors, and benchmark noise.

We present Relational BabyLM, a system submission to the BabyLM 2026 challenge that combines two cognitively motivated inductive biases in a single decoder-only Transformer. Architecturally, we replace standard self-attention with a Dual Attention Transformer (DAT), which separates the routing of object-level ("sensory") lexical features from structural/relational information (Altabaa and Lafferty, 2025; Altabaa et al., 2024; Webb et al., 2024; Kerg et al., 2022; Webb et al., 2021). Relational attention (RA) disentangled from self-attention greatly increases data efficiency and out-of-training-sample generalization on purely relational tasks, but language modeling requires object-level and relational information to be integrated as well as disentangled, and RA-based LMs have remained largely unexplored. BabyLM's data-constrained training and comprehensive evaluation is an ideal testing ground for whether that data efficiency transfers. As a training intervention, we add a Next-Latent Prediction (NextLat; Teoh et al. 2026) objective that encourages hidden states to compress history incrementally into a dense belief state. Architecture is the dominant factor for structural linguistic generalization; the objective is secondary but still significant. DAT's three relational attention types (full RA vs. the simpler RCA and DisRCA variants) are largely interchangeable at 10M words; full RA pulls ahead at 100M. We also introduce a novel symbol-retrieval mechanism (RoPE-based, as opposed to learned, relative symbols) that matches learned symbol libraries while adding no parameters. On the strict (100M-word) track, our best model ranks 6th of 55 overall and 3rd of 55 on the leaderboard's NLP-task subset at the time of writing; our two strongest models outperform the GPT-2 baseline on most benchmarks, with one attaining the highest EWoK score among strict-track entries.

Rigid-body interpenetration frequently occurs in procedurally assembled and generated scenes and must be removed before downstream applications such as physical simulation. We present S4R (Scaling for Rigid-Body Interpenetration Resolution), a scale-continuation method for static interpenetration repair. S4R first uniformly shrinks each body about a fixed reference center to a small initial scale, at which the layout is penetration-free, and then restores full scale through a sequence of minimum-norm convex contact quadratic programs (QPs) that target the linearized separation margin during continuation. Resolution thereby replaces one deep correction with a sequence of shallow-contact subproblems. A conservative scale-event bound and frozen-witness gap predictions cut the number of exact mesh queries; the continuation then ends with a full-scale evaluator check and bounded tail refinement. We evaluate S4R on Kubric, HY3D-Bench, and Thingi10K using a shared mesh-level evaluator and a unified per-scene timing protocol. In the main comparisons on all three benchmarks, up to N=5000 bodies, S4R reaches zero reported penetration with displacement that stays small and nearly independent of scene size, and at the lowest wall time within each hardware tier among the compared methods. A GPU implementation extends these results to large-scale scenes. Our code and data can be found on our project page: https://frank-zy-dou.github.io/projects/S4R/index.html.

As coding agents move from supervised code completion to unattended, around-the-clock exploration, their work expands from isolated predictions into long trajectories of reasoning, tool use, and feedback. Token efficiency therefore becomes important for scaling recursive self-improvement. We take an RSI-inspired approach at the harness layer, scaling auto-research loops across increasingly numerous and diverse environments for harness rollouts. At this scale, the process yields reusable improvements that transfer beyond their development setting, moving automated harness discovery toward production-level outcomes. Four mechanisms survive selection and form SoL-Pi, spanning action execution, context compaction, observation handling, and delegated reading. On the 51-task EdgeBench evaluation, SoL-Pi achieves performance comparable to Pi across GPT-5.6 Sol and Opus 5 while reducing recorded token traffic by 44.7-49.0% and API cost by about one third. In other words, estimated hourly savings are \$8.75-\$13.50 relative to native Codex and Claude Code harnesses, and \$4.36-\$5.71 relative to Pi.

E-commerce livestreams have emerged as an important channel for presenting products to online consumers, containing multiple products whose information is scattered in different moments. This poses significant challenges for downstream product understanding applications, such as product-centric livestream clipping, where models need to identify the product and its relevant segments for information gathering. However, existing benchmarks for general product understanding typically evaluate product retrieval and temporal localization in isolation, leaving the critical correspondence between product identity and temporal evidence largely unassessed. To address this limitation, we introduce GPUB, a large-scale benchmark comprising 3,000 livestream instances with quality-controlled multi-moment temporal annotations and a catalog of over 31K fashion products. GPUB supports three evaluation tasks: the main task Grounded Product Understanding (GPrU) requires jointly identifying the target product and localizing its supporting moments from a livestream video and a candidate product set; Product Retrieval and Product Moment Localization serve as two complementary subtasks. Evaluation of existing multimodal models shows that GPrU remains highly challenging, with the best-performing baseline achieving only 10.13% Pair [email protected]. To narrow the performance gap, we further develop UniPro, a unified product understanding model that derives product-aligned and temporally structured representations from shared multimodal encoding, improving Pair [email protected] to 21.53% while achieving 37.23% Joint R@[email protected] on GPrU.

Rebar insertion is among the most repetitive and physically demanding tasks on construction sites, and a contact-rich problem at 1.4 mm clearance. The parts, however, vary at two levels: a nominal design per structural member, and fabrication tolerance around each nominal design. Real-world data therefore has to be re-collected as designs and batches change. We present RebarSim, a visual sim-to-real system trained entirely in simulation. A privileged state-based teacher is trained with reinforcement learning over procedurally generated rebar geometries, then distilled into a multi-view student that maps raw RGB and proprioception directly to actions under extensive domain randomization. The student transfers to the real world zero-shot, seating rebars taken from a real factory production run in 91.3% of real-robot rollouts. Underlying that result, geometry diversity and pretraining both bring benefits. Training across a diverse set of nominal designs rather than one lifts the zero-shot success of both the teacher and the student on unseen designs, and the student policy outperforms a single-design specialist on that specialist's own design. A pretrained student then adapts to a new design with 4--6x fewer distillation samples than one trained from scratch. Visual sim-to-real transfer depends on appearance randomization and the DAgger mixture: removing either one sharply lowers success. Videos, code, and task assets are available at https://rebarsim.github.io.

Open-vocabulary scene understanding is fundamental for robotics, laying the groundwork for spatial reasoning and object manipulation. While closed-vocabulary 3D instance segmentation heavily leverages 3D shape information, state-of-the-art open-vocabulary methods remain predominantly restricted to 2D image features or image-distilled representations during mask labeling. In this paper, we propose SenseFuse, a label-free fusion method that balances 2D image and 3D shape encoders for robust open-vocabulary 3D instance segmentation, refining only the mask-labeling stage of existing pipelines. We reveal that 2D image and 3D shape encoders exhibit largely disjoint failure patterns and rarely share identical wrong labels, whereas two 2D image encoders frequently repeat the same errors. This distinct behavior makes the 2D and 3D pair inherently complementary. We introduce an adaptive mechanism that selects a scene-level fusion weight to maximize a label-free sensitivity measure, estimated directly from a single scene's unlabeled proposals in milliseconds. SenseFuse improves labeling accuracy in every evaluated setting across ScanNet200, Replica, and ScanNet++, recovering 67-100% (median 93%) of the gain achievable with an oracle weight, and it raises instance AP in 21 of 22 reported settings. Code is available at https://github.com/hanes1207/SenseFuse.

Agent harnesses supply planning guidance, organize execution, and check completion. We study how these components affect success, erroneous acceptance, and cost in two Retail experiments and an Airline pilot in $τ^2$-bench. The primary comparison pairs prewritten task-specific plans (Fixed) with shuffled policy text matched in word count (Sham), isolating the contribution of guidance content. Across 265 matched cells, Fixed improves oracle-verified success by 7.17 percentage points (90\% task-clustered bootstrap interval, 1.15--13.36 points), with gains concentrated in higher-complexity tasks. A read-only terminal verifier rejects 61\% of Retail oracle-invalid episodes while withholding 17\% of correct ones, at less than one cent of additional cost per episode. Which component matters more depends on the loss assigned to erroneous acceptance: at low liability the planning gain dominates; at high liability the verifier's avoided false passes dominate---and a standalone verifier captures nearly all the false-pass benefit of the full planning-plus-verification stack at a fraction of its cost.

Studying the dynamical behavior of a system often depends on characterizing how it transitions between long-lived states. Because such transitions are rare, observing them usually requires specialized enhanced sampling techniques. Transition Path Sampling (TPS) is a well-established method for generating reactive trajectories, which is simple to implement and does not require the definition of a preconceived reaction coordinate. However, its efficiency is limited by its sequential nature and the resulting correlations between sampled paths. Previous work addressed this limitation by combining TPS with a sampling scheme based on conditioned Boltzmann Generators, a generative machine learning model capable of sampling a given target probability distribution. This approach produces uncorrelated transition paths but relies on an accurate reaction coordinate, which is rarely known in advance. Building on recent advances in committor learning, specifically on the Artificial Intelligence for Molecular Mechanism Discovery (AIMMD) method, in this work we introduce GenAIMMD, an iterative algorithm that actively and self-consistently learns the ideal reaction coordinate (the committor) and trains a conditioned Boltzmann Generator to sample from arbitrary bias windows along it. GenAIMMD thereby provides a correlation-free and fully parallelizable path sampling scheme that does not require prior knowledge of the system's transition mechanism. We apply GenAIMMD to a two-dimensional toy model and a higher-dimensional polymer system. In both cases, GenAIMMD succeeds in training the Boltzmann Generator and learning the committor. Benchmark results show a substantial increase in performance compared to standard TPS.

Accurate optimization of a supervised spectral objective need not produce an accurate population subspace or a better predictive representation. We investigate these distinctions for Online Kernel Supervised Principal Component Analysis (OKSPCA), which combines a centered cross-moment in finite random-feature coordinates with an Adam-style orthonormal basis update for an established objective. Fixed-map consistency, concentration and perturbation results describe the estimator and its exact subspace; same-target comparisons then assess the practical iterate separately. Across six predictive benchmarks, performance depends on the declared pipeline: replacing the tracker with the exact empirical target leaves the two regression deficits largely unchanged. Direct classification-rank models capture nearly all terminal objective energy on average, but a saved intermediate state exhibits substantial geometric deviation; a controlled sample-size study further separates empirical accuracy from population recovery. In distinct numerical-service workloads, exact on-request computation is faster in the tested classification settings, whereas Adam saves time relative to the tested full thin-SVD service for some dense wider-regression requests, alongside persistent geometric error. These diagnostics limit explanations based solely on terminal optimization accuracy and distinguish numerical cost from quality, rank coverage and freshness; they establish neither practical-tracker convergence nor predictive or deployment benefits from basis availability.

Numerical site-response predictions often deviate from observations, yet correcting these discrepancies is difficult because records are limited in both sensor coverage and number of events. This study proposes the Transfer-Enabled Forced Latent Autoencoder for Response Equations (FLARE-T) to improve these predictions by learning and calibrating low-dimensional latent dynamics that connect the base acceleration input to acceleration outputs at multiple depths. FLARE-T learns a low-dimensional response manifold and input-driven dynamics from dense finite-element simulations. It then trains a sparse encoder to map simulated sensor responses into the learned coordinates and uses limited records to calibrate the dynamics within them. A short response window initializes each prediction, while the complete base motion drives the response. The framework was evaluated using a layered-soil centrifuge test and the Lotung field vertical array. Test-set results show that FLARE-T improved multi-depth acceleration histories and 5%-damped pseudoacceleration response spectra relative to the original finite-element models, reducing errors at every evaluated sensor for motions of different intensities and, at Lotung, for both horizontal components. Two Lotung source models with different constitutive parameters achieved comparable test-set accuracy, indicating reduced dependence on precise prior calibration. FLARE-T therefore provides a data-efficient means of combining dense numerical response information with limited field records to improve future site-response predictions.

This paper studies resource-limited external influence in time-varying opinion networks when a controller must choose both a small set of agents and a scalar intervention at each update. We use the affine free response, which includes DeGroot and Friedkin--Johnsen dynamics, followed by a direct sparse action. Eliminating the scalar action reduces the one-step problem to a cardinality-constrained support objective whose global optimum is obtained from the largest or smallest residual components. Hence exact sparse one-step-ahead optimal control requires one sort and \(O(|\mathcal T|\log|\mathcal T|)\) operations, where \(\mathcal T\) is the support set. For periodic affine dynamics, the resulting state-dependent low-rank feedback admits cycle-based ultimate bounds and exact tracking of model-invariant targets. The DeGroot and Friedkin--Johnsen specializations expose the role of target invariance and the affine mismatch term. With several competing external players, each sparse best response retains the same sorting structure and the complete binary--continuous stage game is an exact potential game, so a pure-strategy equilibrium exists at every frozen state. Same-state welfare benchmarks separate loss due to scalar competition from the additional loss due to strategic support selection. Numerical examples validate exact support selection and illustrate switching-order effects, DeGroot/Friedkin--Johnsen tracking, resource tradeoffs, and competitive implementations.

Robots searching for a target from a natural-language description must determine not only where to search, but also which observed candidate is the desired target. These decisions reflect two distinct sources of uncertainty - spatial uncertainty over candidate locations and semantic uncertainty over target identity - that are often conflated in VLM-based search systems. We introduce a spatial-semantic uncertainty formulation that maintains separate beliefs over each component and integrates probabilistic VLM evidence into a global target-identity posterior, including probability mass for undiscovered targets. This decomposition allows an information-theoretic planner to independently value candidate discovery and target disambiguation through spatial and semantic expected information gain (EIG), providing an explicit mechanism for trading broader exploration against earlier identification. We evaluate six VLM uncertainty-elicitation interfaces on 500 synthetic targets and show that similar recognition accuracy can conceal substantial differences in calibration and false confidence. In degraded-observation search-and-identify experiments, EIG-based planners reach confident decisions in 75.0%-92.5% of trials, compared with 20.0% for Random search, while different spatial-semantic weightings achieve comparable identification accuracy once confidence is attained. Increasing semantic emphasis reduces unnecessary exploration and VLM queries, demonstrating that explicitly planning over semantic uncertainty can accelerate target resolution without sacrificing decision quality. These results highlight the distinct roles of uncertainty representation and uncertainty-driven planning in embodied VLM systems.

Geospatial foundation models can provide strong flood-segmentation performance, but their size limits deployment on memory-constrained edge hardware. We distill a 300-million-parameter Prithvi-EO-2.0 teacher, fine-tuned on the 252 manually labeled Sen1Floods11 training scenes, into a 0.7-million-parameter EfficientViT-B0 student. The teacher supervises additional unlabeled Sentinel-2 imagery, allowing the student training set to grow without new manual annotations. At the matched budget of 252 scenes, teacher-supervised training is competitive with direct training and improves STURM-Flood performance across tested configurations; a geometry-matched control shows that label source alone does not explain the difference. Scaling the teacher-supervised pool to 2,500 scenes narrows the remaining student--teacher gap: the float student reaches 0.787 water intersection over union on the Sen1Floods11 test split against 0.822 for the teacher, matches the teacher on STURM-Flood under our evaluation protocol, and remains below it on WorldFloods-v2. After activation replacement and quantization-aware training, the student runs as a 1.5-megabyte 8-bit integer (INT8) TensorRT engine on a Jetson Xavier NX at 5.57 milliseconds of graphics processing unit (GPU) compute per 512-by-512 image, with approximately 14 megabytes of runtime device memory. A fixed modified normalized difference water index (MNDWI) threshold is competitive with both models on the two clean external benchmarks, so we interpret those benchmarks as generalization tests rather than as evidence of learned-model superiority over a spectral rule. The results support the conclusion: foundation-model supervision can amplify a fixed manual annotation budget into a substantially larger training set and yield a compact, deployable edge model.

This article presents a mathematical model of the Motivated Emotional Mind cognitive architecture developed for embodied intelligent systems. Such a system learns to maintain its homeostasis through a generalized form of reinforcement learning based on its internal motivations, termed motivated learning (ML). The principal contribution of this article is a rigorous formalization of the re-entrant loop integrating feedforward processing, lateral interactions, and feedback pathways, together with the representational selection mechanisms that govern adaptive system responses. The model specifies how ongoing exteroceptive and interoceptive signals, bodily-motivational context, and memory traces are bound into associative memory structures termed semblions, which compete for access to further processing and top-down reconstruction. The formalization encompasses secondary perception, representational competition, curiosity, procedural gaps, and action selection directed toward limiting allostatic violations. Within this framework, motivated learning is tailored to embodied systems whose dynamics are shaped by needs, affect, and the current regulatory state. Unlike standard reinforcement-learning models, the proposed approach incorporates need thresholds, goal generation and shifting goals, bodily state, resource constraints, and action uncertainty, thereby providing a more adequate account of response selection under regulatory pressure. Global affect functions as a central control signal, modulating the learning rate, representational valence, and the balance between exploration and exploitation. The model presented here is a step toward a more rigorous formalization of cognitive phenomena and may provide a basis for further theoretical analysis, computer simulation, and implementation in artificial-intelligence systems inspired by biological processes.

Automated pain recognition from facial expression could make continuous welfare assessment practical in sheep, but adoption depends on trust: a stockperson cannot act on a score that arrives without justification. We ground a model in the Sheep Pain Facial Expression Scale (SPFES) by letting each detected facial region attend over text embeddings of the clinical descriptors and then test whether the resulting explanations mean anything. They do not. Ablating an entire descriptor changes the predicted logit by about $10^{-4}$, and the most-attended cue agrees with the predicted pain level in only $32.6\%$ of regions, although the attention maps, the learned gate, and the generated text all proposed otherwise. We therefore remove the appearance bypass with a concept bottleneck whose classifier reads only SPFES concept scores, supervised by per-region state annotations that image-level pipelines discard. This costs $0.05$--$0.10$ in Cohen's $κ$ but yields concepts that are demonstrably learned: minority pain-indicating states are recovered at $3.5$--$8.3\times$ their base rates, and the ear and eye severity orderings emerge without severity supervision. Removing the supervision alone leaves $κ$ unchanged while concept accuracy falls to $0.109$, showing that architectural necessity does not imply semantic validity. We also show that pooled concept accuracy is misleading under clinical imbalance and provide a cross-validated, protocol-matched benchmark of seven methods on this dataset.

Document parsing converts document images into structured content and requires reliable performance across diverse layouts and acquisition conditions. Yet training corpora are biased toward common document types and clean digital pages, while expanding coverage alone does not specify how to address a parser's remaining weaknesses. We present WeVisDoc, a two-stage data-centric framework for robust end-to-end document parsing. Stage I broadens semantic, structural, and appearance coverage through heterogeneous data and structure-preserving degradation synthesis. Stage II uses a held-out probe to measure the Stage I parser's residual errors within fixed visual-structural clusters. These diagnostics guide targeted data construction and reallocation of the target-token budget. WeVisDoc-4B achieves an Overall score of 95.38 on OmniDocBench v1.6 and a mean Overall score of 75.54 across the three PureDocBench tracks, ranking first among the compared end-to-end parsers in all four settings. Compared with Stage I, Stage II improves Overall scores for the 2B and 4B models on both benchmarks, with larger gains on the degraded PureDocBench tracks, including a 4.03-point gain for the 4B model on the Real Degraded track.

Graph foundation models (GFMs) aim to learn transferable representations across severely heterogeneous graph domains. However, severe domain shifts in topology, graph scale, and feature semantics impede the construction of a unified, domain-agnostic representation space. To address this, we propose SCGFM-ART, a structure-centric GFM framework that aligns arbitrary graphs onto a shared relational atlas via Amortized Relational Transport (ART). The relational atlas serves as a universal coordinate system defined by a finite set of relational landmarks (bases), while ART directly predicts reusable, end-to-end graph-to-base transport plans, bypassing costly runtime Gromov-Wasserstein optimizations. Under this formulation, SCGFM-ART decomposes a graph into a unified representation: globally via its relational response coordinates relative to the atlas, and locally via its node-to-role structural correspondences. These correspondences project disparate node attributes into a canonical role space, resolving structural and semantic heterogeneity within a singular alignment interface. Rigorously modeling graphs and atlas bases as finite measured relational spaces, we establish coordinate fidelity bounds, prove stability under predicted transport plans, and derive an amortized coverage bound that guarantees our learning objective tightly surrogates ideal relational coverage. Benchmarked across 14 cross-domain graph- and node-level classification tasks, SCGFM-ART achieves state-of-the-art transferability, securing superior average ranks of 2.29 and 1.14, respectively. Topological perturbation analyses demonstrate that node-role transport retains fine-grained structural nuances beyond global coordinates. On real-world benchmarks, the amortized formulation yields 44.2 to 85.1 times faster frozen target-domain inference by avoiding iterative alignment at test time.

Tactile signals provide direct contact and force measurements that are essential for understanding physical interactions and enabling dexterous robotic manipulation. However, tactile sensing requires direct measurement at contact interfaces, making large-scale data collection reliant on intrusive, costly, and restrictive instrumentation. We present TouchSight, a monocular egocentric vision framework for dense full-hand contact force prediction that leverages 500 hours of pressure-glove recordings and extensive hand-object interaction (HOI) data. To address the appearance gap between gloved training data and bare-hand real-world scenarios, we construct TwinTouch-20H: 20 hours of paired visual data in which generative video models re-render gloved recordings as bare-hand observations against new backgrounds while preserving the original measured tactile labels. TouchSight predicts dense force from both gloved and generated bare-hand videos, outperforms prior contact prediction methods on OakInk2, qualitatively generalizes to natural bare-hand egocentric videos from unseen datasets, and improves consistently as glove supervision scales. These results demonstrate that dense tactile signals can be recovered from egocentric vision alone, without tactile instrumentation at capture time.

When aligning frontier models through post-training techniques, it is not possible to directly demonstrate all of the behaviours we want a model to exhibit in all possible deployment environments; our model must generalise outside of the post-training distribution. One proposed solution is alignment midtraining (AMT), which continues pretraining on large volumes of alignment-relevant documents to encourage generalisation in later stages of training. Despite the prominence of AMT as an alignment approach, there is limited public evidence for its effectiveness. To resolve this, we identify several assumptions around midtraining and evaluate them across scale: up to 110 billion-parameter models and 1 billion midtraining tokens. For instance, we study a scenario where post-training data is ambiguous between two possible motivations. We find that midtraining can steer the model's motivation in simple versions of this setting. However, the presence of a tiny fraction of finetuning data which suggests a competing motivation erases the effects of AMT. We also study scenarios in which we want an AI to follow a number of rules, but only demonstrate a subset of them. We find that demonstrations must be present either in midtraining or post-training datasets for these rules to be robustly learned. Based on these and other findings, we do not believe that there is sufficient public evidence for us to confidently state that midtraining can address the core difficulties inherent in aligning powerful AI systems.

Large language models are usually interpreted through concepts that humans already possess: truthfulness, refusal, deception, personality, harmfulness, and related categories. This paper asks whether models may also represent and use distinctions for which no adequate human concept exists. We call such internal structures xeno-representations, and their study xeno-interpretability. We distinguish the human-interpretable semantic space from the xeno-semantic space: the region of model-native representations for which no adequate human conceptual counterpart is available. We show that the space of possible internal distinctions in an LLM is substantially larger than the space available through finite human descriptions. We then separate experimental identification from semantic interpretation: an internal representation may be reproducibly located, geometrically characterized, causally manipulated, and linked to downstream behaviour even when its semantic content cannot be adequately expressed in human terms. On this basis, we sketch an empirical programme to identify xeno-representations. We finally examine the implications for AI safety and multi-agent systems, where model-native representations may propagate and stabilize across interacting agents while remaining only partially visible through human-readable communication. Xeno-interpretability therefore shifts the aim of interpretability from finding human concepts inside models toward discovering and characterizing the representational structures that are native to the models themselves and might affect their behaviour in unpredictable ways.

Robots operating in cluttered environments must often manipulate objects whose locations are only partially observable. A central challenge is deciding whether to acquire another observation or to first manipulate objects that may occlude the target. Conventional task and motion planning (TAMP) approaches typically make this decision using symbolic action costs or expensive geometric planning, neither of which adequately captures how likely an observation is to reveal an occluded target. We introduce Imagine-TAMP, an interleaved planning and execution framework that uses semantic and geometric imagination to compare alternative task-level strategies under partial observability before committing to expensive motion planning. A vision-language model shapes a particle belief over target locations using commonsense relationships between the target and visible objects, while a generative scene model estimates plausible geometry in unobserved regions. Given a target hypothesis and imagined scene, Imagine-TAMP generates multiple symbolic plan skeletons and assigns non-unit costs that approximate both manipulation effort and target visibility from sensing actions, distinguishing a short but poorly informative observation strategy from a longer strategy that first manipulates an occluder to better expose the target. The selected skeleton is then refined into a feasible continuous plan and executed, with new observations updating the belief and triggering replanning when necessary. Experiments show that imagination-guided evaluation improves observation-versus-manipulation decisions: in viewpoint-constrained shelf scenes, non-unit geometric evaluation increases success from 46.0% to 84.0%, while semantic belief shaping further reduces manipulation and replanning. On a real robot, the complete system reduces planning time by 32% relative to a geometry-only ablation.

Offline policy evaluation (OPE) is crucial in high-stakes reinforcement learning applications, where new policies must be assessed reliably before deployment. In such settings, point estimates alone are insufficient; principled uncertainty quantification, such as confidence intervals and variance estimates, is essential for safe and risk-aware decision-making. A comprehensive way to unify these tasks is to estimate the sampling distribution of the evaluation error. Existing approaches, however, often suffer from limited robustness, scalability, or finite-sample validity. In this paper, we propose a model-based bootstrap framework for uncertainty quantification of OPE in finite-horizon, time-inhomogeneous Markov decision processes (MDPs). Unlike classical bootstrap methods that rely on resampling complete episodes, the proposed method regenerates trajectories from an estimated MDP and can therefore accommodate a much broader range of offline data formats, including complete trajectories, transition-level observations, and trajectory fragments. This flexibility further improves finite-sample statistical efficiency. We establish bootstrap distributional consistency, asymptotically valid confidence intervals, and consistent variance estimation for the target policy value. Extensive simulations show that the proposed method accurately captures the sampling distribution of the OPE estimator, yielding tighter confidence intervals and more accurate variance estimates in most settings.

Vision-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to execute long-horizon instructions in unknown environments. Existing zero-shot VLN-CE systems typically maintain spatial states through geometric localization or coordinate-based representations. Recent geometry-constrained navigation removes depth and globally consistent coordinates, but maintaining persistent spatial awareness for place confirmation, progress verification, and recovery remains challenging. We present Navi-Agent, a zero-shot VLN-CE agent that constructs a coordinate-free spatial state from visual observations and executed motion histories. Navi-Agent organizes this state as a navigation topology, where nodes represent visual places and edges represent motion transitions. This representation enables observation-based approximate self-localization, task progress verification, and visual revisitation-based recovery. Navi-Agent performs closed-loop navigation by decomposing instructions into sub-goals, executing local visual navigation, and verifying visited places through the constructed spatial state. Experiments on zero-shot VLN-CE benchmark and real-world robot platforms show that Navi-Agent achieves state-of-the-art performance among geometry-constrained methods while remaining competitive with approaches relying on geometric localization.

Iris presentation attack detection (PAD) is security-critical when a subsystem that appears reliable during development encounters presentation attack instruments (PAIs) or acquisition conditions absent from validation data. We benchmark three compact scratch-trained computer-vision models, each with at most approximately 0.26 million trainable parameters, on the Notre Dame subset of LivDet-Iris 2017 under PAI-driven domain shift and environmental degradation. All models are trained without external pretraining or data augmentation and evaluated over five seeds. A validation-selected threshold is transferred unchanged to the known-attack, unknown-attack, corrupted, and pooled test partitions. From known to unknown attack presentations, Attack Presentation Classification Error Rate (APCER) increases by 17.11-30.47 percentage points and Detection Equal Error Rate (D-EER) increases by 7.38-12.73 percentage points. At the validation-selected threshold, ZACH-ViT obtains the lowest unknown-attack APCER (47.69 +/- 4.84%) and D-EER (38.87 +/- 0.93%), while Compact-TransMIL obtains the lowest Bona Fide Presentation Classification Error Rate (BPCER). ZACH-ViT also gives the lowest unknown-attack BPCER at an APCER limit of 10% (81.29 +/- 1.95%). The high absolute errors show that the comparative advantage of the best compact model does not constitute deployment readiness under unknown PAIs.

We study repeated contract design when a principal observes outcomes but not the actions that generate them. The principal may use any bounded outcome-contingent payment vector, and the agent's best response can make expected profit discontinuous in those payments. For every fixed number $m\ge2$ of outcomes, the minimax regret over $T$ rounds is of order $T^{m/(m+1)}$, up to logarithmic factors. The upper bound allows arbitrary action spaces and agent heterogeneity, without smoothness or monotone-surplus assumptions. Its key is an effective-dimension reduction that the benchmark can be normalized even when fixed tie-breaking is not shift invariant, after which revealed preference yields a monotone response map in payment-difference coordinates. A learning policy built on a Lipschitz parametrization of this map attains the rate using only observed outcome categories. The lower-bound construction accounts for how incentive losses accumulate across outcome dimensions. It shows that each additional contractible outcome creates a precise and unavoidable increase in the worst-case cost of learning.

Large-scale routing often requires visiting clusters of nodes in a prescribed order, giving rise to the Ordered Clustered Traveling Salesman Problem (OCTSP). Optimizing each cluster independently seems natural, but misses non-local dependencies. We introduce the COMPASS algorithm for OCTSP, which combines search with learning-accelerated routing by orchestrating parallel sub-solvers. COMPASS has no quality ceiling and its solutions keep improving with compute. It exploits the clustered structure, and can reach exact solutions in time exponential in cluster size rather than instance size. Empirically, COMPASS consistently outperforms alternative methods. Unlike common large-scale routing solvers, COMPASS consumes general distance matrices and is not limited to coordinate inputs. We demonstrate scaling to 100K synthetic nodes and to 28.5K real e-commerce nodes. To our knowledge, the latter is the largest reported routing solution over asymmetric distances, 9x beyond established ATSP benchmarks.

Spatial reasoning abilities correlate strongly with performance in STEM fields. Games offer a compelling medium for training these critical skills in developing children who have a natural proclivity for play. However, to facilitate human-like tutoring and player guidance, these games require an AI agent capable of making commonsense inferences from spatial events. Qualitative reasoning (QR) models appear to be a suitable framework for these application domains. As these models reason in symbolic representations, they can seamlessly translate game states into interpretable feedback for human-like player guidance. This paper introduces a hybrid qualitative model designed for Camelot Jr., a block-puzzle game that requires constructing multi-level bridges to connect two avatars stationed on separate towers. The game poses a challenge for the player, who must make platforms stable, plan their path, and ensure they use all the provided blocks. To handle the precise physics required by the domain, we integrate a mathematical center-of-mass stability logic to guide our qualitative solver. Our work facilitates spatial skill training in Camelot Jr. and contributes to the development of human-centric, explainable game-playing agents.

LEO satellite networks feature dynamic topologies, time-varying links, and diverse service requirements, which make conventional routing schemes difficult to support fine-grained quality-of-service (QoS) provisioning. Existing studies mainly optimize routing over network states with predefined objectives, but rarely address the practical challenge of translating unstructured natural-language service requests into adaptive routing decisions. To bridge this gap, we propose STR-Agent, an LLM-driven framework for QoS-aware routing in LEO satellite networks. The key innovation of STR-Agent lies in unifying intent perception, tool-based execution, experience accumulation, and reflection-based policy adaptation within a single agent architecture. Specifically, the Perception Module converts natural-language requests into structured routing semantics, while the Reflection Module dynamically adjusts the service-to-routing-policy mapping according to real-time congestion conditions and historical routing outcomes, rather than relying on a fixed routing objective. In addition, we develop a specialized perception model, and construct a domain-specific supervised fine-tuning dataset for LEO service understanding. Simulation results in a Walker-Delta constellation show that STR-Agent significantly outperforms conventional baselines: it reduces end-to-end delay by up to 60% compared with DQ-Dijkstra, improves average intent-understanding accuracy from 45.4% to 92.45% after supervised fine-tuning, and the Reflection Module further reduces the delay by 120 ms at 600 Mbps. These results demonstrate the potential of LLM-driven agent architectures to enable service-aware and adaptive QoS routing in future LEO satellite networks.

Ultrasound image analysis plays a crucial role in cancer screening and prenatal diagnosis, yet comprehensive assessment requires jointly addressing tasks such as lesion segmentation and benign-malignant classification. While recent vision foundation models have shown remarkable universal representations, unlocking their potential for ultrasound is bottlenecked by the considerable domain gap from natural images. Existing methods typically fine-tune heavy vision encoders for isolated tasks, incurring substantial computational overhead while overlooking the underlying commonalities across heterogeneous tasks. In this work, we propose FreqDINO++, a frequency-guided multi-task routing vision foundation model for universal ultrasound analysis. We first introduce a Multi-task Routing Adapter (MR-Adapter) to support parameter-efficient integration of task-common and task-specific knowledge, a Frequency-aware Feature Enhancer (F$^2$-Enhancer) is then designed to capture the rich multi-scale frequency characteristics of ultrasound images, and a Task-aligned Collaborative Decoder (TC-Decoder) is devised to promote collaboration between dense and global prediction tasks through global-local token interaction. Extensive experiments on large-scale multi-task and external single-task ultrasound benchmarks demonstrate that FreqDINO++ consistently outperforms strong baselines and recent foundation models across 27 diverse clinical task scenarios, while also showing promising generalization to unseen data. The code is at https://github.com/MingLang-FD/FreqDINO-Plus.

Traffic regulations are written for human interpretation and therefore rely on shared background knowledge and flexible phrasing, which inherently introduce ambiguity, context dependence, and semantic underspecification. These linguistic characteristics conflict with the precision required by computational reasoning engines such as Prolog, which demand explicit logical structure. This study evaluates two baseline translation approaches, Natural Language to Prolog ($NL\rightarrow Prolog$) and Logical English to Prolog ($LE\rightarrow Prolog$), and introduces a new reasoning-guided translation framework called Structured Four-Stage Legal Translation ($S4L\rightarrow Prolog$). The proposed S4L framework performs semantic role extraction, scene completion, logical mapping, and Prolog rule generation within a single guided prompt, enabling direct translation of raw traffic rules into executable logic without human intervention. A benchmark consisting of twenty real-world traffic rules was used to evaluate each approach in terms of syntactic validity, semantic correctness, and logical completeness. $S4L\rightarrow Prolog$ achieves the highest accuracy, correctly formalizing 75 percent of the rules, while $NL\rightarrow Prolog$ reaches 60 percent and $LE\rightarrow Prolog$ reaches 55 percent. Qualitative analysis further shows that S4L captures implicit causal relations, deontic modality, and exception structure more reliably than the baselines. These results demonstrate that structured reasoning prompts can substantially improve the reliability of natural-language-to-logic translation for legal and safety-critical applications.

Blind and low-vision users often need to locate a specific personal object rather than an arbitrary instance of the same category. The task calls for a robot that can move through the space and reach viewpoints the user cannot, and for an accessible interface where the user says which object is meant and learns whether the right one was found. We present RoboFind, a multi-agent framework in which a smartphone teaches the target and a quadruped robot carries out the search. A Target Teaching Agent converts guided smartphone recordings into a semantic target profile and a reusable multi-view reference bank through an accessible capture flow with AR guidance, speech and haptic feedback, and screen-reader support, so later missions refer to a stored object without repeating the teaching process. At runtime, a Navigation Agent explores the environment and proposes candidate targets, a Verification Agent checks each candidate against the stored references, and a Coordination and Recovery Agent completes the mission or triggers recovery and continued search. Across 32 real-robot missions, RoboFind reaches 85.0% success against 25.0% for a reconstructed sequential first-stop baseline over 20 trials with ten targets, and reduces false success from 75.0% to 5.0%. On six shared targets it succeeds in 10/12 trials, against 5/12 for 12 independently executed GPT-6 Astra-only trials. These results show that the multi-agent design fits the demands of personalized object search, where verifying object identity before declaring completion is what makes the outcome something a user can rely on.

Agricultural image understanding requires fine-grained recognition of plant diseases, pests, crop structures, and botanical species under complex real-world conditions. Despite recent advances in Multimodal Large Language Models (MLLMs), existing models remain limited to text-only outputs and lack pixel-level visual grounding capabilities. In this work, we introduce AgriScope, a unified pixel-grounded multimodal framework for agricultural image understanding. AgriScope jointly supports image-level, region-level, and pixel-level understanding within a unified framework, enabling tasks such as grounded caption generation, referring expression segmentation, and multi-turn multimodal interaction for agricultural imagery. AgriScope integrates biologically specialized semantic representations with dense spatial grounding through biological-semantic encoding, dense spatial representations, and pixel decoding. To support large-scale grounded learning, we introduce AgriGround, a large-scale pixel-grounded agricultural multimodal instruction-tuning dataset containing over 500K images and 11M instruction-following samples spanning plant disease analysis, crop and weed identification, insect pest recognition, and fine-grained botanical understanding. AgriGround is constructed through a multi-stage automatic annotation pipeline that integrates multimodal caption generation, phrase-level grounding, segmentation mask generation, and task-oriented instruction synthesis to produce densely grounded supervision. Extensive experiments across multiple agricultural vision-language tasks demonstrate the effectiveness of AgriScope in pixel-grounded multimodal understanding, establishing a strong benchmark for agricultural vision-language learning and visual grounding. The dataset and code will be made publicly available at (https://github.com/boudiafA/AgriScope)

Three-dimensional reconstruction from unorganized point clouds remains a challenging problem in computer vision, geometric modeling, and computer-aided design. While neural implicit methods achieve impressive reconstruction accuracy, geometry is typically encoded in latent representations that limit interpretability and reuse within engineering workflows. We present NeuSOGA3D (Neuro-Symbolic Geometric Abstraction in 3D), a hybrid framework that combines learned perceptual priors inherited from NeuSOGA with explicit symbolic geometric reasoning. The method projects point clouds onto principal orthographic planes, constructs symbolic implicit spline representations from the resulting observations, and fuses them through shape-preserving constructive solid geometry operations to generate a coarse visual hull. Additional geometric detail is recovered through cross-sectional decomposition and volumetric reconstruction using Partial Shape-Preserving Splines. Unlike conventional neural implicit approaches, NeuSOGA3D progressively transforms observations into explicit symbolic entities, including control polygons, implicit spline fields, cross-sections, and volumetric lofts. Experiments on all forty categories of the ModelNet40 benchmark demonstrate the ability of the framework to recover structurally meaningful and CAD-compatible geometric representations from diverse point-cloud observations. The results highlight the potential of combining learned perception with symbolic geometric reasoning for explainable geometric intelligence.

Ubiquitous time series data across diverse domains enables critical applications in areas such as transportation systems and power grids. Recently, training foundation models on massive datasets to achieve accurate zero-shot forecasting has emerged as a major research focus. However, current studies predominantly prioritize architectural innovations while insufficiently addressing data diversity, often relying on simple data sampling strategies that fail to manage complex data distributions effectively, leading to inefficient use of training data and suboptimal performance. To address this, we propose QUALS, a large-scale time series corpus equilibrium framework. QUALS significantly enhances data efficiency, i.e., enabling existing models to achieve superior performance using only a small fraction of the original training data. Specifically, QUALS operates through two core mechanisms. First, a pattern quantization framework systematically decodes heterogeneous patterns from mixed corpora via vector quantization and uniform binning. Second, a learnability synchronization framework calibrates sampling weights for heterogeneous patterns, bridging the optimization gap between simple and complex motifs to maximize overall training efficiency. Extensive benchmarks demonstrate that pre-training on QUALS consistently achieves superior zero-shot performance, even under substantially reduced training budgets.

Generally, most voice agents are cascaded systems, i.e., an ASR model transcribes the caller's audio, a language model reads the transcript and decides what to say and which backend tools to call, and a TTS model speaks the reply. Nearly all of the decision making happens in the language model, but existing evaluations measure it either too broadly or too narrowly. End-to-end voice benchmarks score the full pipeline, so recognition errors and model errors mix into a single number. LLM benchmarks isolate the model but they do not evaluate what makes real phone calls hard, such as transcription issues, caller's voice being split across messages and the requirement that replies follow the language and script specified. We introduce the Multi-Turn Voice Agent Benchmark (MTVA-Bench), which evaluates the language model on the same conditions it faces inside a cascaded system. The caller is played by an LLM following a set of rubrics and tool calls are answered by a mock backend which responds to the arguments the model actually sent. The benchmark contains 49 agents working across 490 reviewed scenarios and supports 7 languages. Scoring is a combination of deterministic checks on tool calls with two LLM judges, one that scores scenario specific rules and one that grades conversation quality without access to the task. Both judges must cite specific messages from the transcript. Task and conversation scores are weighted equally, since a call can complete its task and still go badly for the caller. In a seven-model study, six of the models select the correct tool within 6.4 points of one another, but their overall scores span 24.4 points. Most of the gap comes from argument values, action ordering, rule compliance, and what the model says around its tool calls.

This work proposes a physics-enhanced machine learning approach for the system identification of Linear Time-Varying (LTV) systems under time-varying operating conditions in terms of fast-varying natural frequencies and damping ratios by combining a long short-term memory network with an Extended Kalman Filter (EKF). The proposed approach uses vibration data (displacement and velocity measurements), domain knowledge of modal damping ratios, and a physics-based model that can yield an approximate natural frequencies time-dependency model. The approach is validated using synthetic data generated from a finite element model of a 2-blade offshore wind turbine under realistic environmental and operating conditions. This system displays fast time-varying frequencies due to operating conditions, whose identification is particularly challenging because of the wind and wave loading. The robustness of the proposed approach is assessed under assumed incorrect system information (e.g. damping ratio). The proposed approach is evaluated across different environmental and operating conditions to show its applicability to different operating regimes. The results show the approach can accurately identify the selected fast-varying natural frequency, 1st Fore-Aft (FA-1) mode, with a maximum root mean square error of 0.0012 Hz. The results demonstrate that the model trained on EKF estimates depends on accurate damping values, whereas the model trained on physics-based data exhibits robustness to incorrect damping assumptions. The approach is extended to damping ratio identification for the selected mode by estimating the root mean square error between models trained on EKF estimates and physics-based data. The results show that the approach can yield a good approximation of the FA-1 mode damping ratio using grid search, offering an improvement over covariance-driven stochastic subspace identification.

Reasoning-based reranking with Large Language Models (LLMs) has shown promising improvements in text ranking. However, current methods predominantly rely on a single reasoning trajectory, resulting in rankings that are susceptible to reasoning errors and inherently constrained in modeling the multifaceted signals underlying document relevance. To resolve this dilemma, we propose MERIT-Rank(Multi-perspective Evidence and Reasoning Integration for Text Reranking), a framework that models complementary reasoning trajectories to improve reranking robustness. MERIT-Rank formulates a Multi-Trajectory Reasoning Space (MTRS) that evaluates query-document relevance from multiple perspectives and introduces a joint reranker that consolidates these reasoning paths into a unified ranking decision. We further develop Progressive Rank Policy Optimization (PRPO), a progressive training framework that stabilizes reasoning trajectories while continually improving ranking quality through staged optimization objectives. Experiments on both reasoning-intensive and traditional retrieval benchmarks show that MERIT-Rank consistently achieves superior performance over competitive baselines. The 4B model notably outperforms most 7B and even 32B rerankers on BRIGHT.

Recent memory-augmented repository-level program repair methods reuse historical repair experiences to improve LLM-based issue resolution. However, our analysis reveals three limitations in existing repository-level memory retrieval. First, episodic memory is highly imbalanced across repositories, leaving low-resource repositories with little effective support. Second, more memory does not monotonically lead to higher repair success, suggesting that relevance, quality, and redundancy matter more than raw memory volume. Third, memory accumulation is phase-misaligned: repositories may contain many reproduction experiences but few patch or refinement experiences. To address these problems, we propose an adaptive experience retrieval framework for repository-level program repair. Our framework introduces coverage-aware retrieval, which falls back to cross-repository or repair-type-based memories when same-repository memory is insufficient; quality-aware selection, which ranks memories by relevance, historical utility, specificity, and redundancy; and stage-aware routing, which separates and retrieves memories for reproduction, localization, patch generation, patch refinement, and validation. Evaluated on SWE-Bench-Lite and SWE-Bench-Verified, the proposed framework improves repair performance on under-covered repositories, reduces noisy memory retrieval, and better supports failed-to-fixed patch refinement. Our results show that the key to memory-augmented repair is not simply accumulating more experiences, but retrieving the right experiences for the right repair context.

Evaluating live streaming speech synthesis (TTS) requires assessing fine-grained, highly expressive prosody such as emotion, intonation, and energy which traditional MOS predictors fail to capture. While proprietary Large Language Models (LLMs) like Gemini can evaluate these aspects, they are too costly for massive inference and reinforcement learning feedback. To address this, we first introduce Live-ProsodyJudge (LPJ), a cost-effective pairwise evaluator distilled from Gemini into Qwen3-Omni. However, we identify a critical flaw in standard multi-dimensional evaluation: verdict coupling. The judge tends to lazily align all individual dimension scores with its overall preference, collapsing a rich multi-dimensional rubric into a single preference bit. To resolve this, we further propose Decoupled-Live-ProsodyJudge (D-LPJ). D-LPJ eliminates the overall verdict target to prevent blind following, masks uncertain pair-dimensions during Supervised Fine-Tuning(SFT), and introduces a novel span-local GRPO strategy that applies normalized advantages strictly to their corresponding rationale spans. Evaluated on highly curated human-annotated test sets, 10 sample balanced-order LPJ achieves higher point accuracy than a single Gemini call, while D-LPJ successfully produces independent,decoupled dimension judgments. Furthermore, in a Best-of-8 TTS candidate selection tournament, the LPJ-selected utterance falls within the human top-3 in 85.29% of high-confidence cases, demonstrating its efficacy for fine-grained TTS preference optimization.

We study GPT-6-Astra in a zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) system, where it interprets instructions, assesses its surroundings, and proposes actions. The system uses a common observation--decision--execution workflow with direct model API calls, without a packaged agent harness or navigation-specific fine-tuning. In this workflow, each request receives selected observations, execution feedback, and retained progress records. Evaluation covers the complete system, including context management and action control. We evaluate the system on 50 of the 100 R2R-CE val-unseen episodes used by Open-Nav. It achieves a success rate of 52.0\%, an SPL of 48.9\%, and an nDTW of 70.8\%. Our analysis highlights three findings. First, recorded responses link landmarks and earlier actions to instructions using observations and supplied history. Second, reviews include requests for additional views and revisions of uncertain judgments. Third, the results suggest a gap between task understanding and autonomous completion: an unfinished crossing is recognized while rotation continues. At termination, 36.0\% of episodes succeed with a workflow-accepted STOP, while another 16.0\% meet the distance criterion at the step limit. These results highlight a central challenge: translating correct local judgments into sustained progress and appropriate stopping.

Collecting real-world navigation data for mobile robots typically requires platform-specific teleoperation, making large-scale collection expensive and difficult to scale. We introduce Universal Navigation Interface (UNI), a robot-free data collection paradigm that uses a four-wheeled rollator walker (rollator) and smartphone to collect physically constrained human demonstrations. Because the rollator cannot climb stairs, negotiate uncut curbs, or pass through narrow gaps, demonstrations are naturally biased toward wheeled-feasible routes. Using UNI, we collect 37.2 km of real-world navigation data and recover metric trajectories that directly supervise goal-conditioned navigation models. Fine-tuning visual-navigation models on UNI reduces trajectory prediction error by 17.4-24.8% on held-out UNI demonstrations. Evaluation on other navigation datasets shows benefits that vary by dataset and metric. We further demonstrate closed-loop transfer to a powered wheelchair in curb, staircase, and curb-cut scenarios. These results support low-cost physical proxies as a practical source of navigation supervision collected without the target robot.

Visuomotor policies for multi-fingered dexterous manipulation are highly sensitive to camera viewpoint shifts. To achieve view invariance, recent methods increasingly rely on explicit 3D modalities like RGB-D or point clouds, which can introduce hardware dependencies, calibration requirements, and vulnerability to sensor noise during real-world deployment. In this work, we show that view-invariant control can be achieved without explicit test-time 3D sensing by encoding geometric knowledge into the visual representation during simulation. We present AnyViewDex, an asymmetric training pipeline that combines multi-view contrastive alignment with privileged 3D geometric supervision. By regressing absolute 3D object coordinates during simulated training, this auxiliary objective provides a geometric grounding signal that mitigates the spatial collapse of the globally pooled contrastive embedding. At deployment, the policy operates zero-shot using only uncalibrated monocular RGB and proprioception. We validate this approach across both reinforcement learning and student-teacher distillation. In hardware evaluation on an xArm7 with a 16-DoF LEAP Hand, AnyViewDex reaches 76.7% grasping success across eight unseen objects and six uncalibrated viewpoints (480 trials; 2,400 across all ablation conditions), indicating that geometrically grounded monocular policies transfer zero-shot without test-time depth. Project Page: https://anyviewdex.github.io/

Robotic reward models evaluate task execution from visual observations, but their predictions can change with camera viewpoint and occlusion even when the underlying task state is unchanged. Adapting a pretrained reward model to a local task therefore requires accounting for how that task is observed. We introduce AnyviewMeter, a geometry-conditioned adaptation framework for robotic reward models that represent task progress as a scalar reward signal. It combines low-rank fine-tuning with token-aligned Plucker rays and synchronous block attention: ray conditioning incorporates camera geometry into visual features and attention queries and keys, while block attention fuses synchronized views inside the pretrained decoder. The framework supports both single-view reward prediction and joint multi-view evaluation through parameter-efficient adaptation of a pretrained Robometer model. On PickCube, single-view adaptation improves progress prediction in every camera group and reduces mean absolute error under a changed field of view by approximately 21% relative to RGB fine-tuning. Across simulated manipulation tasks, joint multi-view prediction reduces progress error by 41-69% compared with averaging single-view RGB predictions and improves temporal ordering in approximately 88% of task-camera groups. On real tasks with fixed and wrist-mounted cameras, mean absolute error decreases by approximately 21% relative to averaged RGB fine-tuning. These results support camera geometry and joint visual evidence as useful components of task-specific robotic reward adaptation.

We describe the architecture, training methodology and inference speedups of Granite 5.0 Turbo CTC, a 470 million parameter encoder-only model with an excellent speed-accuracy tradeoff. The architecture uses pyramidal temporal subsampling within Conformer blocks using strided depthwise convolutions, block-diagonal (chunk-wise) self-attention, and conditioning on intermediate predictions from the middle layer. Training highlights are the use of only publicly available data, the novel use of a Muon optimizer, and balanced data sampling. Inference speedups include replacing 1 x 1 convolutions with linear layers and optimizing the attention computation in the Conformer blocks. Collectively, these result in a model that is on the speed-accuracy Pareto frontier of the Open ASR leaderboard for English short-form ASR while being twice as fast as the fastest competitor. The model can be used under a permissive license and downloaded from https://huggingface.co/ibm-granite/granite-speech-5.0-470m-turboctc.

Reliable temporal-difference targets are central to off-policy actor-critic learning. Direct value improvement refines the next-state target with alternative actions, but the reliability of this refinement depends on how candidate actions are ranked, reviewed, and weighted. Noisy rankings may force premature candidate commitment, reusing selection scores may bias target valuation, and fixed enhancement weights may amplify weak evidence. To address these risks, we develop Conservative Adaptive Ranking and Screening (CARS), which retains an ordered candidate prefix within a preset budget and narrows it only when the observed boundary gap exceeds a disagreement-scaled uncertainty radius. Selector-Evaluator Value Assessment (SEVA) uses selector critics to order candidates and a separately parameterized evaluator critic to review the selected value, then caps the reviewed value at the selector reference. Dynamic Adaptive Risk-aware Enhancement (DARE) then regulates each residual correction using candidate reliability, the gap between selector and evaluator signals, and a finite stage factor. Together, CARS, SEVA, and DARE form CARE-VI, an evidence-regulated target construction framework that preserves the backbone interfaces for critic regression and actor updates. The analysis bounds the CARS boundary error, the SEVA selected-value overestimation, and the one-sided deviation of the DARE residual displacement from its population counterpart, and establishes fixed-policy recovery after the finite-stage perturbation ends. Experiments with SAC, TD3, and TD7 on four MuJoCo tasks show that CARE-VI achieves the highest mean return in all twelve settings. Grouped ablations and scalar diagnostics support the roles of the three components in improving target reliability.

The Internet of Agents is expected to enable large numbers of autonomous agents to discover, verify, and collaborate with each other across heterogeneous platforms. However, current agent protocols mainly address tool invocation and inter-agent communication, leaving scalable agent registration, trustworthy identification, and capability-oriented discovery largely unresolved. To address this, this paper proposes a scalable trust discovery architecture for the Internet of Agents. The proposed architecture adopts a hierarchical and distributed design consisting of three layers: Agent Root for trusted registry governance, Agent Registry for agent registration and metadata publication, and Agent Resolver for distributed capability discovery and trust-aware resolution. The architecture further introduces a registry-suffix-anchored composite identity scheme, which binds an agent native identifier to a trusted registry suffix to generate a globally discoverable identity. It also incorporates a dual-certificate and multi-level authentication mechanism to strengthen identity trust among agents. We implement a prototype and evaluate it through large-scale agent registration and resolution experiments. The prototype achieves an average registration latency of 58ms and an average discovery latency of 25ms, and it supports more than 19,000 registration requests per second and more than 29,000 agent discovery requests per second. These results demonstrate the feasibility of the proposed architecture, providing a practical approach toward scalable and identity-trusted agent ecosystems in the Internet of Agents.

The Antibandwidth and Cyclic Antibandwidth problems are NP-hard graph labeling problems that aim to maximize the minimum (cyclic) distance between labels assigned to adjacent vertices. Extensive research on these problems has resulted in a variety of mathematical formulations and computational approaches. However, their minimum span perspective, in which a prescribed minimum (cyclic) distance is fixed and the objective is to minimize the label span, has received comparatively little attention. In this paper, we consider this complementary perspective by introducing the Minimum Span Antibandwidth/Cyclic Antibandwidth Labeling (MSABL/MSCABL) problems and developing a unified Boolean Satisfiability (SAT)-based framework for solving them. The SAT-based framework formulates MSABL/MSCABL as a sequence of decision problems and exploits their monotonicity to accelerate the search process. We also consider two SAT solving strategies, parallel and incremental SAT solving: the former examines multiple candidate spans concurrently, while the latter reuses a single SAT instance while progressively restricting the label domain. The proposed approaches are evaluated on benchmark instances from the Harwell-Boeing Sparse Matrix Collection and compared with CPLEXCP, CPLEXMIP, and Gurobi. The results show that SAT-based approaches are highly competitive in solution quality, with the parallel approach performing best overall for MSCABL and the incremental approach for MSABL. With the no-hole constraint, they remain competitive with CPLEXCP and significantly outperform CPLEXMIP and Gurobi, particularly for MSCABL. These results demonstrate the effectiveness of SAT solving as an exact approach for MSABL and MSCABL.

Self-evolving methods reduce the need for human-annotated trajectories by allowing tool-using agents to generate their own training data. Yet existing methods typically separate trajectory generation from evaluation, relying on static verifiers that cannot adapt to emerging failure modes or self-consistency signals that may reinforce errors shared across trajectories. Jointly adapting planning, execution, and evaluation offers a promising alternative, but introduces a fundamental coordination challenge: each component continuously changes the data or feedback used to train the others. We address this challenge with \textbf{UnifiedPlayers}, a cooperative framework comprising a Planning Player that generates tasks, an Execution Player that produces multi-turn trajectories with Python tool calls, and an Evaluation Player that constructs executable verifiers. We design role-specific rewards that coordinate the three players toward a shared learning objective under GRPO. Across two model backbones and twelve reasoning benchmarks, UnifiedPlayers outperforms the strongest prior baseline by at least 3.5\% on mathematical reasoning and 3.9\% on general reasoning tasks. Moreover, the learned verifier achieves 84.2\% adversarial detection accuracy, while its reward signal exhibits 2.03$\times$ higher per-question variance than a self-consistency baseline, providing more discriminative verifications. These results highlight cooperation among specialized players as a promising path toward self-enhanced tool-integrated agents.

Effective manipulation across diverse objects is critical for applications ranging from agricultural harvesting to laboratory and domestic automation. While the inherent compliance of soft robotics is well suited to this challenge, designing grippers that generalize across tasks remains difficult due to the vast design space of continuum mechanics and the risk of overfitting to specific scenarios. We propose a graph-based design space for representing soft structures and mechanisms, coupled with a multi-objective, diversity-driven genetic optimization framework that explicitly promotes solution variety throughout the design process. Using multiple grasping scenarios during optimization, we study how task diversity influences the emergence of generalization to unseen objects and contact conditions. Our results show that optimization over a sufficiently diverse set of grasping cases leads to designs with emergent generalization, exhibiting improved robustness compared to task-specific solutions on novel scenarios. These findings suggest that diversity-driven optimization offers a principled pathway toward general-purpose soft grippers, aligned with the adaptable nature of soft robotics.

Long-term multivariate time series plays a significant role in many application areas such as power systems, trading, etc. However, their accurate prediction is quite difficult for conventional forecasting methods as they often exhibit high dimensionality and complex relationships. Recent works show that transformer-based approaches are quite effective for long-term forecasting thanks to their attention mechanism. However, in the presence of complex high-dimensional inputs, they show evidence of oversmoothing, limited capacity, and opacity. To this end, this paper introduces SETTer, a transformer-based model that addresses these challenges by incorporating novel techniques for decoupled self-attention and hybrid masking. The proposed techniques enable SETTer to effectively capture the dominant short- and long-term patterns across the temporal and channel dimensions. In addition, we enrich the model layers with simple explainable structures that indicate the discriminative pattern of SETTer. We show that with a single-layer transformer architecture, SETTer can effectively model long-term dependencies in the presence of varying data complexities. Extensive experiments on real-word benchmark datasets for long-term multivariate time series forecasting demonstrate that SETTer outperforms state-of-the-art models in 88% of the scenarios.

Tool learning enables large language models (LLMs) to use external tools for tasks beyond parametric knowledge. Reinforcement learning can optimize tool-call behavior from feedback, but current methods still face two problems: fixed-threshold curricula can become misaligned with the policy's evolving capability boundary, and additive rewards can leak argument-level credit when the predicted tool is wrong. To address these problems, we propose MATCH, a closed-loop framework for model-aware tool learning with curriculum scheduling and hierarchically gated rewards. Model-Aware Curriculum Learning (MACL) maintains reward-derived sample difficulty that co-evolves with the policy, and each epoch selects samples near the current capability boundary together with a top-k pool of harder cases. Hierarchical Tool-call Gated Reward (HTGR) scores tool name, argument key, and argument value as a gated chain, granting credit at each level only when prerequisites hold. The same HTGR rewards drive both GRPO updates and MACL's difficulty refresh, closing the loop between policy optimization and sample scheduling. On API-Bank and BFCL V3, MATCH reaches 72.19% and 62.87% overall accuracy, outperforming the main supervised and RL-based baselines. Backbone experiments further show consistent improvements across four backbones from two model families.

The growing complexity of multi-domain operational environments (land, aerospace, naval, cyber, and electromagnetic spectrum) has increased the volume and velocity of data reaching command-and-control (C2) centers, straining the observe-orient-decide-act (OODA) decision cycle. Artificial Intelligence (AI) systems currently employed in defense are, in general, reactive and isolated tools that still rely heavily on human operators to integrate information, assess scenarios, and formulate courses of action. This paper proposes a conceptual Agentic AI architecture for AI systems that can plan, access data sources, execute tools, and act autonomously and audibly, aimed at supporting decision-making across the three Brazilian Armed Forces (Navy, Army, and Air Force). Four application fronts are discussed (decision support, situational analysis, feasibility studies, and countermeasure suggestion), as well as the data and sensor access requirements and the security and permission safeguards necessary for responsible employment across administrative, strategic, operational, and tactical contexts.

Active visual exploration of tabletop objects often requires reorienting an unknown resting object onto a different stable support face to expose occluded surfaces. To identify such placements without exhaustive physical search, we learn a probabilistic placement prior from a single-view point cloud. Stable placement prediction is inherently multimodal, and conventional 6-DoF regression introduces further ambiguity by modeling translation and in-plane yaw. We therefore propose FlipToSee, a probabilistic framework that removes this representational ambiguity by parameterizing placements as unit support normals on $S^2$ while modeling their multimodal conditional distribution via a von Mises--Fisher mixture density network. To decouple mode diversity from physical robustness, FlipToSee deterministically extracts a compact candidate set from the mixture components and applies robustness-aware reranking using an auxiliary head trained with candidate-aligned supervision. In simulation, FlipToSee achieves $98.4\%$ first-proposal success on in-distribution objects, $95.3\%$ on out-of-distribution shapes, and $90.0\%$ under zero-shot transfer to household YCB objects. We further demonstrate the learned placement prior on a physical robot by integrating it with grasp and motion planning for exploratory regrasping.

This paper builds a theoretical bridge between the economic notion of marginal utility and two machine-learning constructs, matrix factorization and the Key--Value cache of transformer language models. The singular value spectrum of a rating matrix is shown to be a diminishing marginal utility schedule for latent factors, the eigenvalue spectrum of the projected covariance operator to be the marginal utility schedule of a model's learned representation, and cache eviction and low-rank cache compression to be instances of constrained utility maximization under a memory budget. The three collapse into a single allocation rule: retain the top dimensions whose eigenvalue exceeds the shadow price of the binding constraint. The framework is applied to the automated extraction of structured information from geo-mining documents, where it motivates a multi-pass inference protocol, a layer-wise TIES model merging procedure, and a selection policy combining extraction quality, localization drift and energy, scalarized with a Conditional Value-at-Risk term on drift. Two empirical contributions are reported. An 11.2-million-parameter hierarchical classifier, trained in about five minutes on a single GPU, reaches 90.0 per cent level-1 accuracy on a held-out test set from a 973-document uranium-exploration corpus, against 92.0 per cent for a proprietary model on a fifty-document human audit of the same corpus, at a latency of 2.62 ms per card against approximately 2,000 ms for the API and at negligible cost. A diagnostic of uniform-density TIES merging exposes a reproducible degenerate mode in which the merged model returns token-identical outputs across five geographically distinct districts while declaring high confidence; re-executing the merge under layer-wise calibrated densities removes that signature on the diagnostic sample. The full-scale extraction benchmark, including LoRA fine-tuning, is reported as projected rather than measured and remains an empirical extension of this work.

Deep learning models for multi-modal breast cancer diagnosis achieve high predictive accuracy but remain clinically unacceptable without actionable, counterfactual explanations. Attribution-based methods (LIME, SHAP) are categorically inapplicable to this purpose, as they generate no alternative instances and thus cannot be evaluated on counterfactual quality metrics. This investigation provides empirical evidence that FCA-Guided Counterfactual (FCA-CF) framework that uses a Formal Concept Analysis (FCA) concept lattice as a hard structural constraint on counterfactual search, operating over a multi-modal TCGA-BRCA dataset. We benchmark against four genuine counterfactual methods: Wachter-style CF, DiCE, FACE, and NICE, evaluated on 60 benign-predicted TCGA-BRCA instances. The FCA-CF framework achieves Validity = 1.0000 (100% of counterfactuals successfully flip the prediction), Sparsity = 2.37 features changed (best among all valid methods), and Proximity = 0.900 (normalised L2-based, matching NICE as joint best). The classifier achieves Accuracy = 0.980, F1 = 0.976, ROC-AUC = 0.9947. Ablation analysis confirms that the FCA lattice constraint is the primary sparsity driver (removing it increases sparsity by +40%, p < 0.001, Cohen's d = 0.78), while Phase C greedy refinement accounts for the largest individual contribution (+113% sparsity increase when disabled, p < 0.001, d = 5.01). FCA-guided counterfactual generation achieves a clinically important Pareto-dominant outcome; it is simultaneously the sparsest and among the most proximate of all valid methods, with perfect validity. The emergent sparsity property arising from lattice topology rather than numerical penalty terms constitutes a structurally novel contribution to the counterfactual explanation literature.

Event cameras offer high temporal resolution and motion sensitivity for tiny UAV detection, yet distant targets generate sparse and fragmented events that are easily overwhelmed by clutter and ego-motion. Existing methods mainly rely on dense event representations or local sparse spatiotemporal modeling, resulting in redundant computation or fragmented modeling of motion continuity across distant asynchronous events. To address this limitation, we introduce serialized motion evidence accumulation, which treats motion continuity as an ordered evidence propagation process. Specifically, the same event stream is organized into locality-preserving spatiotemporal paths and chronology-preserving temporal paths through the latent complementary serializations. Based on this principle, we propose PointEvent, a lightweight event-wise state-space framework that alternates serialized scans across the complementary orders, progressively consolidating fragmented motion evidence beyond fixed local neighborhoods. A high-resolution event branch preserves fine-grained target responses, while compact context modulation suppresses interference. Experiments demonstrate that PointEvent achieves SOTA with the fewest parameters and fastest measured inference among the compared methods. Code: https://github.com/wzz-z/PointEvent

The rapid advancement of speech synthesis and voice conversion technologies has made audio deepfakes increasingly realistic, posing serious security risks in practical applications. While existing detection methods achieve strong performance under controlled conditions, they often fail to generalize under real-world perturbations and corruptions. In this paper, we propose ROGUE, a framework that dynamically constructs robust detection workflows by orchestrating multiple detection tools. ROGUE formulates workflow generation as a sequential decision-making problem and introduces a dual-agent paradigm, where a perturbation agent generates audio perturbations and a policy agent learns to select and execute detection tools under perturbed conditions. Through adversarial learning, ROGUE enables perturbation-aware tool selection, adaptive execution strategies, and improved robustness to distribution shifts. Extensive experiments across multiple datasets and real-world corruptions demonstrate that ROGUE consistently outperforms strong baselines in both robustness and generalization. Our results highlight the effectiveness of adversarially optimized workflow generation for building reliable audio deepfake detection systems in real-world deployment settings.

Because of its collaborative nature, Wikidata suffers from errors, in- consistencies, and excessive complexity, such as redundant classes, ambiguity between instances and classes, wrong taxonomic paths, and type constraint violations. The manual curation of these issues is infeasible at scale. To address these challenges, we introduce WiCleanData, a refined version of Wikidata with a consistent tax- onomy and free from type constraint violations. Specifically, we have designed an automated pipeline that first cleans the taxonomy with language model assistance, then simplifies type constraints by hierarchical aggregation, and finally filters facts accordingly. The resulting knowledge graph, free from any type violation, is made publicly available via a Web interface, enabling easy exploration and downstream applications.

Hierarchical robotic systems executing long-horizon manipulation tasks must make high-level semantic decisions that orchestrate stochastic low-level skills. In this setting, failed rollouts are ambiguous: a poor downstream state may reflect an invalid high-level decision, partial observation, or a valid decision whose physical execution failed. Traditional supervised learning lacks data for such recovery states, while reinforcement learning struggles with sparse rewards and non-local credit assignment. We propose MAGMA-GEN, an on-policy data-generation pipeline that converts ambiguous failed rollouts into validated recovery supervision. MAGMA-GEN first uses a privileged coach to hypothesize an early decision-level error and propose localized correction or recovery actions. Because this diagnosis is fallible, candidates are retained only if re-execution from the same state under matched conditions improves downstream progress. This produces supervised examples from the agent's own failure distribution without per-step human demonstrations. Evaluated on interactive long-horizon manipulation tasks, MAGMA-GEN improves task success and recovery capabilities, against distillation and trajectory-repair baselines under evolving task constraints in both simulation and real-robot execution.

A coding agent halfway through an issue has already read much of what a retriever ranks highest. Relevance is scored per passage, but sufficiency belongs to the set: a ranker can fill its budget with variants of one required fact and leave the decision unsupported. We formulate state-conditioned minimal sufficient evidence recovery: given a captured agent state, recover a compact evidence combination that supplies the support its next decision still lacks. SERBench measures this on 500 held-out states from 45 repositories, recording what the agent has seen and crediting only sets that cover every fact the current decision was annotated to require. MSS-Complement treats acquisition as set construction, not ranking. Three semantic calls propose a jointly sufficient set, search for what it lacks, and return 4-8 intact source units within 6,144 tokens. One configuration, fixed on calibration data, recovers a complete set for 73.0% of those states at five items and 80.6% at eight, against 61.4% and 72.4% for Qwen3 embedding with reranking. A matched control ranking by similarity alone reaches 66.6%, placing the gain in the set-level policy, not the computation. From frozen repository source with no gold-derived pool, the lead is 5.0 points. On AMA-Bench it answers from a 76.2% smaller answer prompt, with accuracy 2.08 points above that benchmark's own memory agent. Removing one required group from an otherwise complete set costs 12.3 and 11.1 points of repair-localization precision under two executors. Retrieval for agents is better posed as recovering what a decision lacks than re-ranking what an issue resembles.

Herbicide-based weed control is increasingly unsustainable due to rising weed resistance and the adverse environmental impacts of chemical use. While mechanical weed control avoids these drawbacks, it is typically implemented using large machines that cause soil compaction. We propose a novel alternative based on small mobile robots for mechanical weeding. Compared with existing automated mechanical weeding approaches, the proposed method offers reduced soil compaction, simpler automation, and improved scalability. Our solution involves a Boston Dynamics Spot quadruped robot equipped with a custom weed removal tool featuring a milling bit at its end. The tool is rigidly attached to the robot and uses the degrees of freedom of the robot base by actuating the legs, while keeping the feet stationary. We develop a software architecture that enables autonomous weed removal and integrate this system with all other required components. We analyze the accuracy and efficiency of the current proof of concept both in an indoor and outdoor environment and provide recommendations for future work to make the system more accurate and efficient.

A memory can answer a current query correctly while discarding distinctions required by a later update. We investigate this failure with a paired-history audit: two histories have the same current answer, receive a shared future update, and require different subsequent answers. A pilot evaluates 24 history pairs across six synthetic mechanisms, 12 memory conditions, two repeats, and two model backends. A deterministic frontier selector obtains strict reveal accuracy of 96/96 on DeepSeek and 82/96 on GLM; a structured writer obtains 62 successes with one unresolved outcome and 56/96. The configured four-outcome joint contrast has finite-sample identification intervals of [0.521, 0.542] and [0.292, 0.313], not confidence intervals. A record-level audit distinguishes retained-state adequacy, response delivery, and answer-schema compliance without changing those original scores. It finds 26 and 25 well-formed but semantically wrong structured reveal memories, while all 14 GLM frontier reveal failures contain correct values in the wrong wrapper. Tombstone removal produces 16/16 exact replay failures in the targeted mechanism. Identifier renaming then exposes a separate flaw: original frontier late-reference adequacy falls from 8/8 to 94/320 transformed instances. We provide and test a label-equivariant repair, but it preserves only 2/8 original late-reference answers: eliminating a naming shortcut does not solve unknown future relevance. These results support a scoped evaluation methodology and reproducible failure analysis, not general superiority of the repaired algorithm. Paid pilot evidence, retrospective diagnostics, and new offline tests are reported separately; no independent held-out or natural-task validation is claimed.

This paper presents dynamics-relaxed model predictive control (DR-MPC), a novel MPC formulation for legged locomotion, and a tailored interior-point method (IPM) solver. The formulation combines online optimization feasibility by construction with a contact-aware input parameterization. DR-MPC moves the dynamics equality and affine input constraints into quadratic penalties and retains only nonempty box constraints. The resulting box-constrained quadratic program (QP) has a block-arrow Hessian that enables the state and affine-output directions to be eliminated through a Schur complement. The solver factors only the reduced control system after swing-force elimination and contact-aligned move blocking. For the evaluated implementations using the same DR-MPC formulation, our method achieves median end-to-end MPC speedups of $16.0\times$ over HPIPM and $4.4\times$ over OSQP, with comparable locomotion performance in simulation. DR-MPC achieves a median onboard MPC end-to-end time of $4.4$ ms and is validated on a Unitree Go1 quadruped. Open-source code will be made available after publication.

We present Astronex-World 1.0, an open controllable video world-model foundation. Given a text prompt (text-to-video) or an initial observation (image-to-video), the model predicts future visual states under frame-aligned camera trajectories, continuous actions, and an embodiment identifier, and accepts text events inserted at a specified position of a rollout. The family provides a bidirectional model for full-context generation and a causal model with block-causal attention and cross-block KV caching for persistent generation, both built on the Wan2.2-TI2V-5B prior. PRoPE injects camera intrinsics and extrinsics, while a 64-dimensional action stream modulates every Transformer layer. A five-stage training path develops bidirectional camera and action control, converts the backbone to block-causal generation, distills a few-step student, restores mixed-domain dynamics, and applies asymmetric DMD/DMD2 distribution matching. The causal model generates 832x480 video at 24 fps. All five training stages run on two NVIDIA L20 48 GB GPUs, and the causal model streams in real time on one. It scores 73.5 on WBench Navi and 70.0 on WBench Full. On Full, this 5B model is above the 13.6B LongCat-Video and the 14B Helios, within one point of the 22B LTX-2.3, and above YUME 1.5, which is post-trained from the same 5B prior on NVIDIA A100 GPUs. The reserved action input and output interfaces allow post-training for embodied intelligence and autonomous driving.

Urban traffic forecasting often relies on information distributed across stakeholders who may be unable to share raw data due to privacy or commercial constraints, motivating federated spatial-temporal approaches. In such federated settings, each client observes traffic over a distinct sensor subgraph with its own spatial topology and temporal dynamics, leading to significant heterogeneity across clients. Existing federated spatial-temporal methods typically rely on model parameter aggregation and provide limited mechanisms for recovering spatial dependencies across client boundaries. This introduces two key limitations. Specifically, parameter aggregation across heterogeneous graph domains tends to dilute client-specific representations, while road network partitioning breaks the propagation of traffic dynamics across client boundaries. To address these challenges, we propose FedeRICo, a federated traffic forecasting framework that combines gradient-level collaboration with boundary-aware residual communication. FedeRICo employs a dual-branch forecasting architecture in which a globally guided branch captures transferable forecasting structure, while a private residual branch preserves client-specific corrections and incorporates boundary residual signals. The global branch is coordinated through gradient alignment across all clients, enabling collaborative optimisation without destructive parameter interference. To recover cross-client spatial dependencies, boundary messages are extracted through a trend-residual decomposition that suppresses periodic structure and communicates only transient spatial-temporal residual signals between physically adjacent clients. Experiments across four real-world traffic forecasting benchmarks demonstrate that FedeRICo consistently outperforms state-of-the-art federated spatial-temporal baselines while maintaining competitive training runtime.

The EU AI Act (Regulation 2024/1689) imposes technical obligations on high-risk AI providers, yet Articles 8-15 were drafted for predictive AI and leave seven technical gaps when applied to generative systems, spanning non-deterministic data governance, training-data provenance, continuous conformity, human oversight, open-ended robustness, emergent risk, and generative fairness. We deliver Governance-as-Code (GaC), a framework of 43 machine-checkable acceptance criteria across six compliance modules that run in a CI/CD pipeline and emit Article-indexed audit evidence, and we show the actual Rego policy code rather than merely describing it. Our central commitment is that the Act's open-textured standards ("appropriate levels," "possible biases") become declared, auditable numbers: robustness thresholds are derived from the provider's documented baseline and a state-of-the-art floor, and framing bias is collapsed into eight measurable proxies tested by counterfactual demographic probing. We also correct who owes what, since under Article 25 and Chapter V a downstream deployer relies on the upstream provider's Article 53 training-data summary and documents only the layers it controls, so GaC verifies that summary rather than demanding per-sample documentation the deployer never had. We validate on two enterprise deployments, a high-risk advisory chatbot and a limited-risk content generator, benchmarking against a manual expert audit rather than documentation artifacts that were never designed to enforce compliance. GaC reproduces all of the manual audit's findings, including three penalty-triggering violations, while cutting audit labor by roughly 75%.

Feed-forward 3D reconstruction provides an efficient paradigm for scene modeling from image sequences. Scaling these models to large monocular scenarios are constrained by excessive GPU memory footprint, degraded local geometry, and long-term trajectory drift. Existing chunk-based optimization strategies provide limited geometric constraints and fail to maintain global consistency over extended trajectories. We present a unified framework for stable and scalable feed-forward 3D reconstruction from long monocular sequences. Our approach builds on coarse-to-fine trajectory alignment augmented by lightweight geometric prior injection. Distilling monocular geometric cues into the feed-forward backbone via LoRA adaptation improves depth accuracy on fine structures while preserving inference efficiency. We introduce a hybrid-weight sparse ray-field optimization that leverages high-frequency geometric features to guide local point-cloud refinement and enforce consistent inter-frame ray constraints. Unlike prior chunk-based methods, this establishes strong cross-frame geometric coupling while maintaining scalability. Finally, an efficient trajectory stitching strategy with joint ray-error optimization explicitly reduces accumulated drift. Extensive experiments show that our approach achieves competitive trajectory accuracy compared with representative SLAM systems, while maintaining globally consistent 3D reconstruction in large-scale scenarios.

Omni models can describe video content, but can they locate events in time, preserve event order, and judge audio-visual synchronization? We introduce AVTrace (Audio-Visual Temporal Reasoning Assessment and Capability Evaluation), a silver-standard diagnostic suite spanning onset and span grounding, synchronization, next-step prediction, cross-modal localization, chain parsing, and event-conditioned comprehension. It contains 34,114 training examples and category-balanced development and test splits of 3,500 and 7,000 examples. We evaluate five open omni models under their respective input configurations using reference-blind response normalization followed by deterministic scoring. All five off-the-shelf systems score below the test split's majority-label baseline of 0.556 on synchronization verification, and obtain low scores on chain parsing and event-conditioned grounding and comprehension. Development-set perturbations reveal task-dependent sensitivity in Qwen3-Omni-30B to modality removal and changes in visual input processing, without isolating their underlying causes. Parameter-efficient temporal post-training improves Gemma4-E4B-it on several benchmark metrics. On three external image benchmarks, task metrics change modestly, including some degradations, while teacher-forcing perplexity decreases. Together, these findings show that semantic reference-text overlap should not be treated as a proxy for temporal localization, and that AVTrace can identify task-specific weaknesses while providing a testbed for temporal post-training.

The widespread adoption of LLMs has led to escalating content compliance risks. Prior works have contributed to addressing these risks in the English context, downplaying the complexity of Chinese language content. This paper follows China's current AI-Generated content compliance requirements and provides evaluation results on 20 notable LLMs, offering insight into China's regulatory landscape. We design a novel framework to assess the compliance and refusal rates with 2303 questions spanning six distinct dimensions, including 203 self-constructed constitutional questions. The framework employs several judges to generate verdicts independently based on their hierarchical alignment memory. Our findings show that international models also exhibit high levels of compliance despite the use of standard Chinese questions, and the main differences may stem from dimensions closely related to ideological alignment. We establish a regulatory benchmark that enables the global AI community to evaluate both Chinese and non-Chinese LLMs under a unified set of legally grounded compliance requirements.

Graph Neural Networks (GNNs) achieve state-of-the-art performance on graph-structured data, but training and inference on large graphs are often bottlenecked by memory constraints and sparse linear-algebra workloads. Quantum computing offers an alternative set of primitives that may improve scalability for graph learning. Building on the quantum graph neural network (QGNN) framework of Liao \textit{et al.}, this work implements two representative architectures --- the Simplified Graph Convolution (SGC) and Linear Graph Convolution (LGC) models --- and evaluates them on open benchmark graph datasets and semi-supervised learning tasks using quantum simulation. We compare predictive performance and optimization behavior against classical baselines, showing that the quantum models achieve competitive performance with fewer parameters. Finally, we present a cost gradient analysis that identifies the tasks for which the models showcased are trainable. This is followed by a classical simulability study to find regimes in which the proposed circuits remain robust during training.

Artificial intelligence (AI) and neural networks have driven groundbreaking innovations across numerous disciplines. Optical computing offers the promise of unprecedented speed and energy efficiency in the post-Moore era; however, achieving efficient, practical nonlinear activation using all-optical approaches remains a challenge. Here, we present an optical nonlinear neural processing unit (ONNPU) that implements all-optical nonlinear activation through a self-rectification mechanism. The ONNPU architecture perfectly imitates the structure of digital neural networks, enabling seamless integration with the established deep learning ecosystem. We benchmark ONNPU across nine diverse tasks spanning decision, regression and generation, including accuracies of 98.07% on MNIST and 93.54% on Fashion-MNIST. When integrated into a 201-million-parameter Vision Transformer, ONNPU achieves 82.4% top-1 accuracy on full ImageNet classification (1,000 categories); when integrated into a 117-million-parameter decoder-only Transformer, ONNPU enables short-form story generation that outperforms GPT-2. By addressing more complex and diverse deep learning tasks, ONNPU paves the way toward practical optical machine intelligence, unleashing significant potential for high-performance optical computing.

Vision-language-action models tell a robot where to move, but not how hard to push. Contact-rich tasks depend on that second quantity, compliance, yet no widely used demonstration interface records it. The obstacle is identifiability as realized pose and measured force cannot separate the operator's intended equilibrium from their stiffness, so VR controllers, SpaceMouse and handheld grippers cannot supply compliance supervision even in principle. Prior compliance-output policies work around this with hand-specified task structure, privileged simulation contact state, or dedicated force and tactile hardware. Four-channel bilateral teleoperation removes the ambiguity directly by using the leader arm as a separate measurement of the intended equilibrium, making per-axis stiffness identifiable by regression using only the joint-torque sensing already on the manipulator. This yields per-timestep, direction-dependent compliance labels at zero annotation cost, which we use to fine-tune a VLA to emit stiffness alongside pose. On a Franka Research 3 wiping task, ours is the only policy of five whose contact force changes when the instruction asks for a firm wipe rather than a normal one (6.4N (normal) to 9.1N (firm) RMS, Cohen's d = 0.89, p = 0.023

Long-horizon robot manipulation requires not only stable local visuomotor control, but also continuous target tracking and reliable task progress assessment throughout execution. This challenge becomes particularly critical when multiple objects share identical appearances and must be manipulated in a prescribed order. In such scenarios, relying solely on a limited-horizon manipulation policy is often insufficient to determine which instance should be operated on and when the task should transition to the next stage. To address this challenge, we propose MaskHarness-WAM, an instance-grounded harness for long-horizon manipulation. The proposed system connects high-level task planning with low-level manipulation policies through target masks, while leveraging visual feedback for subtask scheduling and continuous execution. Since each subtask corresponds to a different target instance, the low-level policy requires a newly established initial target mask under the updated scene at each subtask transition. The harness continuously re-observes the environment, generates, and verifies the target mask at subtask boundaries, thereby updating the instance-level spatial condition provided to the low-level policy. Furthermore, the system advances the manipulation process by switching target instances according to the verified completion status of each subtask. Experiments on a real robot platform demonstrate that MaskHarness-WAM substantially outperforms limited-horizon policies on sequential multi-object manipulation, showing its effectiveness in extending local manipulation skills to reliable long-horizon execution.

Event cameras provide low-latency, high temporal resolution perception for real-time vision tasks such as robotics.The novel circuitry (i.e. asynchronous, independent pixels) that enables these advantages also introduces new algorithmic challenges. Velocity-invariant representations alleviate missing observations under slow motion and motion blur under fast motion, but most discard temporal information by converting events into image-like representations. We propose Set of Centre Active Receptive Fields (SCARF), a real-time velocity-invariant representation that preserves raw events while consistently handling fast motion, stationary scenes, and independently moving objects. SCARF achieves state-of-the-art performance in both computational efficiency and representation quality.

The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation, prefill remains computationally expensive, and large KV caches continue to strain HBM and SSD capacity and data-transfer bandwidth. Together, these compute, storage, and bandwidth demands constitute the primary bottleneck to further lowering deployment costs. To address this challenge, we introduce DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts (MoE) model with 552B backbone parameters and support for contexts of up to one million tokens. With its Causal Encoder-Decoder (CED) architecture, the model activates 16B parameters per token during decode but only 8B parameters during prefill, substantially improving cost efficiency for agentic workloads. To push the limits of KV cache compression, DeepSeek-V4.1-Flash combines cross-layer KV cache reuse in Compressed Sparse Attention 2 (CSA2) with FP4 KV caching. These designs reduce its global KV cache footprint (always in HBM) to 890 bytes per token, roughly 1/4 of the corresponding footprint of DeepSeek-V4-Flash. Further, through a dedicated deployment optimization known as SWA Bounded Replay, DeepSeek-V4.1-Flash reduces its persistent KV cache footprint (always on SSD or in host memory) to roughly 1/8 of that of DeepSeek-V4-Flash. Despite its much smaller KV cache footprint, the model delivers substantially better performance than the baseline. In addition, we streamline the DeepSeek-V4 architecture and introduce several efficient architectural extensions. We pretrain DeepSeek-V4.1-Flash on a multimodal corpus comprising 45T tokens and conduct comprehensive post-training, yielding strong performance across diverse text-based and multimodal agentic scenarios. Model checkpoints are available at https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash.

Large Language Models (LLMs) are increasingly applied to legal and criminal justice tasks, yet existing work focuses almost exclusively on post-arrest scenarios where the suspect's identity is already known, leaving the critical pre-arrest challenge of inferring suspect characteristics from incomplete evidence largely unexplored. To fill this gap, we introduce the Profiling, Investigation, and Judgment (PIJ), comprising 2,500 real homicide cases from five countries. PIJ evaluates LLMs across three tasks that span the entire criminal investigation pipeline: criminal profiling, which requires abductive reasoning to infer suspect attributes from fragmentary scene evidence, crime process reconstruction, which tests structured information extraction, and sentence prediction, which demands legal deductive reasoning. We evaluate 9 powerful LLMs and find that performance degrades systematically as tasks shift from explicit fact extraction to implicit reasoning over unknown suspect profiles. Categories requiring inferential reasoning, such as motivation and victim-offender relationships, remain the primary bottlenecks. Further analysis reveals substantial gaps between LLMs and human experts, along with pervasive biases in gender, age, and motive attribution. Our findings indicate that pre-arrest inference from incomplete evidence remains an open challenge.

Placing a grasped book into a tight shelf is a compact but difficult contact-rich control problem: millimetre-scale pose error can turn a geometrically valid approach into jamming, failed release, or incomplete seating. We study this final phase after grasp acquisition and global approach, and ask how control authority should be divided between known geometry and learned behaviour. Our method retains a nominal task-space controller for structured insertion and seating, while residual PPO supplies bounded local corrections and decides when to release. Only the brief open-retreat-reclose transition is scripted. For the final policy used on hardware, a deployment-matched simulation evaluation over 512 fixed conditions yields 98.50 percent mean success (0.23 percentage-point sample SD) across three independent training runs, compared with 37.89 percent for nominal control. On the physical xArm7, 60 trials over 30 matched conditions show the same qualitative advantage: residual control raises success from 26.7 percent to 63.3 percent, reduces failures from 22 to 11, and wins 13 of the 15 matched conditions in which the two controllers differ. Robustness tests show that performance remains above 87 percent under initialization perturbations up to 1.5x, while very tight clearances expose the geometric limit of local correction. These results support a hybrid design in which geometry preserves reliable task structure and learning is concentrated on the contact-sensitive behaviour that fixed rules handle poorly.

Networked low-altitude unmanned aerial vehicles (UAVs) need reliable and adaptive decision-making capabilities to operate under uncertain observations, dynamic environments, and intermittent connectivity, while many existing agentic systems remain limited by hallucination risks, data dependence, and weak generalization. This article investigates neuro-symbolic agentic AI (NSAAI) as a framework for combining neural grounding, symbolic reasoning, and closed-loop agentic interaction to support more reliable and adaptive UAV autonomy. We first examine its capability foundations in data efficiency, compositional generalization, continual learning, and zero-shot transfer, and then develop a reference architecture integrating task and goal management, neuro-symbolic planning, verification and metacognition, skill execution and network interaction, and shared knowledge and memory. An urban fire-inspection case implemented in LAESim illustrates how a UAV can coordinate sensing and cloud access under intermittent connectivity, reuse a verified image-delivery skill, and satisfy explicit evidence conditions before completing the mission. The results illustrate the potential of NSAAI to support reusable skills, evidence-grounded decision-making, and adaptive mission execution in networked UAV systems. We further discuss key research directions in uncertainty-aware reasoning, knowledge and skill expansion, adaptive self-monitoring, and standardized evaluation.

Tree-based Monte-Carlo Tree Search (MCTS) duplicates the same state when it is reached through different trajectories, which can waste simulations in stochastic MDPs. We introduce Graph-Based Stochastic-Power-UCT (GS-Power-UCT), which shares states reached at the same planning depth while keeping separate values for states reached at different depths. This design applies to general stochastic MDPs, including problems with cycles. We prove that for a fixed planning horizon, the root estimate converges to the finite-horizon value at rate $O(n^{-1/2})$, matching tree-based Stochastic-Power-UCT while reusing samples across shared states. We also study two full-state variants: GS-Power-UCT-F, which stores one node per physical state to increase sample sharing but may mix values from different remaining horizons, and GS-Power-UCT-F$^+$, which uses an adaptive horizon to control this bias. The latter converges to $V^{\star}(s_0)$, the optimal infinite-horizon discounted value at the root state $s_0$, when the remaining cross-depth gap vanishes. Experiments on stochastic planning benchmarks show improved sample efficiency over tree-based and graph-based baselines.

This work proposes a real-time 6 Degree-of-Freedom (6 DoF) tracking algorithm for monocular laparoscopic surgery. It provides alignment between intra-operative video and pre-operative data (e.g., CT). The 6 DoF tracking offers a solution for accurately locating the internal anatomy of the target organ despite the lack of tactile feedback and transparency. The ORB-SLAM2 framework is adopted and modified for prior-based 3D tracking with four major modifications. First, the primitive 3D shape is used for fast initialization of the ORB-SLAM2 monocular mode. Second, a pseudo-segmentation strategy is employed to separate the target organ from the background for tracking. Third, the 3D shape is incorporated as a geometric prior in its pose graph optimization. Fourth, the Multi-Scale Retinex with Chromaticity Preservation (MSRCP) algorithm is leveraged and modified for image enhancement in challenging illumination scenarios. In-vivo and ex-vivo experiments validate that LapaTrack-3D provides robust 3D tracking and effectively handles typical challenges such as poor illumination, fast motion, out-of-field-of-view scenarios, partial visibility, and ``organ-background'' relative motion. LapaTrack-3D achieves a processing rate of 13 Hz for 1280*720 pixel video.

LLM-based AI agents process user requests through iterative reasoning and tool execution, often involving the invocation of remote LLM APIs with local tool containers. This execution model can make the optimization of agent serving difficult because latency, local resource demand, and container bottlenecks inter-mix across requests. However, the current agent ecosystem runs without much consideration of resource dynamics, which results in significant waste of the precious resources. This paper analyzes the resource inter-mix of AI agents for three representative tasks: retrieval-augmented question answering, web search, and software coding. To this end, we characterize the latency with respect to the resource dynamics of processing multiple requests and tasks concurrently. Our measurements show that agents have a wide range of behaviors depending on tasks, so that even the same tool can differ substantially in resource dynamics. We also find that running multiple requests concurrently exposes task-dependent bottlenecks in resource dynamics such as CPU, disk I/O, and memory. Furthermore, we uncover that faster LLM responses or more CPU cores do not always accelerate agents. Based on these observations, we demonstrate new optimization opportunities that exploit the resource dynamics of tasks: CPU-aware tool admission and task-aware CPU allocation. Our results show that the latency of CPU-sensitive agent tasks improves $\sim$5.4$\times$, and the average latency across multiple tasks is reduced $\sim$32% compared to native agents.

A geometrically valid placement can still be difficult to execute because the selected grasp changes the required end-effector pose, collision geometry, and transport motion. Placement is formulated as a pre-execution ranking problem in which supplied grasp-placement candidates are scored before planning. The model combines a typed target-conditioned point cloud with three pose descriptors and hierarchical heads for planning success and execution success conditioned on planning. On a 30-object, 1,235-scene dataset with scene-group-held-out splits, three-seed top-1 success on covered test groups reaches 85.63 +/- 1.08% for joint selection and 79.84 +/- 0.16% for fixed-target ranking. For the designated frozen seed-42 checkpoint, top-1 success improves from 72.84% to 85.78% over full-pool cuMotion for joint ranking and from 59.65% to 79.67% for fixed-target ranking. Frozen transfer to xArm7/MoveIt requires no xArm-specific retraining. Across 27 locked cases, 13 complete end to end (48.15%). Of the 16 cases that pass Top-5 preflight and begin execution, 13 succeed (81.25%). Candidate-level deployment-feasibility prediction reaches 81.25% recall, 85.20% specificity, and 83.23% balanced accuracy.

Large language models (LLMs) have been applied to causal discovery, but candidate-graph generation rarely treats premature omission of potentially relevant causal relations as an explicit design objective. We propose MaSCoD, a multi-agent framework that organizes candidate third variables and local structural patterns before direct-edge judgment. We evaluate MaSCoD on Auto-MPG, DWD, and Sachs using GPT-5.4 as the primary backbone and GPT-4o for replication. MaSCoD exhibits a dataset- and backbone-dependent retention-selectivity profile rather than uniform superiority. Across all six dataset-backbone settings, Full, which supplies structural hypotheses before direct-edge judgment, achieved higher mean Recall and F1 than No Phase 1, which instead constructs them within the judgment procedure, while also increasing false-positive rates. Additional reference-edge retention over all evaluated baselines was observed on DWD with GPT-5.4 and on Sachs with GPT-4o, rather than uniformly across settings. Partial ablations showed that supplying both information components did not always outperform supplying only one. For GPT-5.4, stage-wise analysis showed that the Full-No Phase 1 retention gap was already present after direct-edge judgment, while reconciliation introduced additional reference-edge loss for Full on Sachs. These findings support structural pre-organization as an explicit design and evaluation target for omission control and motivate evaluating context construction jointly with its utilization in judgment.

Vision-language-action models (VLAs) have emerged as a promising paradigm for general-purpose robot learning, with performance improving as models and datasets scale. Scaling robot data collection, however, remains challenging because data are naturally distributed across robots, tasks, and locations, making centralization costly or impractical. Federated learning offers a way to train on decentralized robot data, but applying it to VLAs requires accounting for heterogeneous robot client data distributions. We present Co-VLA, which applies consensus optimization using the Alternating Direction Method of Multipliers~(ADMM) to federated VLA training. We show that the same algorithm supports both full-model training and parameter-efficient fine-tuning with both fixed-rank and rank-adaptive adapters. The name Co-VLA reflects both consensus and collaboration: clients with different local robot datasets collaboratively train a shared model without sharing their data. Our experiments demonstrate that Co-VLA achieves performance comparable to centralized training in both full-model training and parameter-efficient fine-tuning settings.

Large language models (LLMs) are typically evaluated on static benchmarks, even though natural language constantly evolves through newly emerging words and meanings. Existing Korean benchmarks are centered on established vocabulary and therefore provide limited coverage of such recent lexical change, and their English-oriented design makes it difficult to assess the typological properties of Korean, in which content words combine productively with functional morphemes. In this paper, we introduce KoNeoBench, a benchmark for evaluating LLMs' understanding of Korean neologisms. KoNeoBench is built on 1,785 Korean neologisms attested in online news since 2020 and curated through expert lexicographic review. Each entry provides usage examples, word-formation analyses, and dictionary-style definitions. Based on this resource, we define four tasks and report results on recent models, together with a human baseline. Our experiments show that current LLMs exhibit clear limitations in recovering source components, distinguishing semantic categories, and generating accurate definitions. These results reveal specific aspects of recent Korean lexical change that remain challenging for current LLMs. KoNeoBench is available at https://github.com/bcmilab/ko-neobench/ .

The study of opinion dynamics in social networks is one of the key challenges in computational social science with direct relevance to understanding political polarization, misinformation, and health responses. Current approaches focus on simplified mathematical models that ignore linguistic and contextual factors related to belief updates or use Large Language Model (LLM)-based simulations that have not been validated against real data. We present a framework based on the concept of a digital twin to simulate opinion dynamics in social networks. The approach fills the gap by cloning a real-world Twitter network, assigns a set of attributes for agents (such as persona, emotions, centrality, stubbornness, and influence), and employs Mistral-7B to perform opinion update based on memory and social exposure. To evaluate the proposed approach, we validate it against two real Twitter datasets (COVID-19 discourse and U.S elections 2020). The results show that the capability of the proposed framework reproduces opinion trajectories and reduces individual prediction error by more than 50% compared to the best-performing classical baseline (Mistral-7B achieves Mean Absolute Error (MAE) = 0.150 and 0.121 on the COVID-19 and US Election 2020 datasets, respectively). We observe similar improvements in structural alignment (Delta_r = 0.120 and 0.180) and polarization dynamics (Delta_Var = 0.106 and 0.115) on the two datasets, respectively. Additionally, the ablation studies confirm that agent attributes, memory, and social exposure all contribute to the framework's predictive fidelity in reproducing opinion trajectories, with agent attributes being the most critical contributor. Overall, our results demonstrate that grounding Mistral-7B within empirically cloned interaction networks produces a realistic simulation framework capable of reproducing complex social dynamics.

Teams of mobile robots rely on continuous communication with their neighbors for coordination, yet most distributed model predictive control (DMPC) schemes assume the communication network stays connected rather than actively enforcing it. Adding such a guarantee is hard since the usual mathematical condition for connectivity is nonconvex and links every agent to every other, which is incompatible with a scalable distributed real-time controller. We propose a DMPC framework in which each agent is assigned a connectivity contract: a local region prescribing where its predicted positions may lie over the prediction horizon. The contracts are designed so that, as long as every agent stays within its own contract, the team is guaranteed to remain connected. Given the maintained contract graph, an agent builds its contract from a single exchange with its immediate neighbors, after which every agent solves its own optimization problem independently. We prove that the resulting closed-loop system maintains connectivity, avoids collisions, and respects local state and input constraints. Simulation and hardware experiments on miniature autonomous car-like robots demonstrate the approach.

3D grounding aims to localize target entities in complex scenes from natural language and plays a fundamental role in embodied perception and spatial reasoning. However, existing approaches mostly rely on feature similarity or direct matching, making it difficult to connect natural-language intent with the implicit semantic and geometric structures hidden in billion-scale urban point clouds. We reformulate city-scale 3D grounding as structured constraint reasoning, where description semantics are organized into computable cross-modal constraints over open-vocabulary 3D entities, attributes, and spatial relations. We present CitySTAR, a training-free framework for reasoning-driven urban 3D grounding. CitySTAR lifts raw billion-scale urban point clouds into a query-ready scene graph of open-vocabulary 3D instances, with CodeLLM-driven tools supplying multimodal evidence for node attributes and 3D spatial relations. It then models target-context topology with paired hypergraphs and performs bidirectional topology verification for structural disambiguation. Finally, a Reflective Cross-modal Grounding module integrates topology consistency and candidate-centered 2D visual evidence to make decisions over a metric-aware 3D context graph. To further support this setting, we introduce CitySTAR-3D, an enhanced benchmark that improves semantic coverage, instance completeness, bounding-box fidelity, and spatial-relation complexity in city-scale 3D grounding. Extensive experiments show that CitySTAR consistently improves open-world urban 3D grounding while maintaining strong interpretability and generalization.

Closed-loop robot policies require observation processing, state management, and situation-dependent branching, making them costly to design and tune manually. Although coding agents increasingly support control-code generation and optimization, it remains unclear whether implementations improved on source tasks also support policy acquisition for new tasks. We study this question by treating complete closed-loop implementations as reusable execution experience. For each source task, a coding agent generates policy code from a few successful demonstrations and iteratively improves it using simulation feedback. The validation-selected implementations are retained in a software archive. For new tasks, the agent generates and improves policies using archived implementations, target demonstrations, and execution feedback. The resulting policy is then frozen and executes without further model calls. Across four source tasks in RoboCasa, iterative optimization increases mean success from 28.3% to 64.2%. Across nine target tasks and three independent runs, mean success is 45.2% without references, 41.5% with initial source code, and 57.0% with optimized source code. Optimized references outperform initial references in all three runs on the nine-task average, with a mean gain of 15.6 percentage points. These results demonstrate the value of execution-improved software as a resource for acquiring new policies in this setting, although initial references remain better on two target tasks when averaged across runs.

We study an in-silico substrate in which every pixel of a two-channel cellular-automata grid carries a tiny neural network (an agent) that senses its Moore neighborhood. A cell persists only by self-replication: a living neighbor is cloned and its weights are mutated by a uniform perturbation, so that phenotype (cell state) is driven entirely by genotype (network weights). From a handful of seeded founders the system grows into a spatially organized ecosystem of coexisting, competing and dominating species. Our main contribution is a battery of coarse-grained diversity metrics that make such growth measurable at two scales: four phenotypic tools based on cellular-type frequency, entropy and cell variance, and two genotypic tools that colour each agent by a hash of its full weight vector versus a sparse random-weight probe. Across a five-fold sweep of 1680 small runs and 24 long (1000-generation, 200 x 200) runs, the substrate is persistent and self-maintaining in 20 of the 24 long configurations and exposes a clear phenotype-genotype diversity trade-off: raising phenotypic diversity collapses genotypic diversity and vice versa. Full-genome hash colouring further reveals lineage structure that a random-weight probe systematically misses. Code, data and animations are released as supplementary material.

Static malware detection for Windows Portable Executable files demands a careful balance between detection effectiveness, computational efficiency, and analytical interpretability. This paper introduces Delphi Scanner, a static malware detection system for Windows PE files that balances efficiency with behavioral interpretation. It uses a convolutional neural network (CNN) to model Windows API sequences to classify PE and a decoupled interpretation layer based on a rule-based layer to categorize APIs into high-level malicious capabilities. Evaluated on over 190,000 Windows PE files, the system achieves 95.35% accuracy with a 1.53~MB model footprint. Robustness experiments on 5,647 out-of-distribution MalwareBazaar samples, paired packed and unpacked executables, and three adversarial manipulation strategies confirm generalization beyond the training distribution and resistance to functionality-preserving evasion techniques. Overall, these results demonstrate that API sequence-based static analysis offers a practical, interpretable, and efficient foundation for malware triage in local deployment scenarios.

Historical archives pose a difficult retrieval problem for retrievalaugmented generation systems: documents are OCR-degraded, heterogeneous across genres and sources, and require strong source traceability for scholarly and institutional use. We introduce TRACE, a training-free agentic retrieval framework designed for accountable source discovery over historical corpora. The system was developed in the context of DECIDON, an interdisciplinary project on the circulation of political discourse between parliamentary debates and the press during the French Third Republic, involving digitised historical collections and institutional use cases. The prototype is currently deployed internally within the project and accessible to 24 researchers across six partner institutions. We evaluate TRACE on HistoriQA-ThirdRepublic, a benchmark of 1,752 French historical questions over parliamentary debates and newspapers from 1887, with documents derived from Biblioth{è}que nationale de France digitised collections. TRACE achieves R@10 = 0.856 and MRR = 0.653, outperforming sparse, dense, graph-based, and agentic RAG baselines, with the largest gains on multi-hop and cross-corpus questions. At approximately $0.02 per question under the default hosted inference configuration, TRACE also remains economically feasible for heritage institutions, laboratories or companies that cannot rely on costly local GPU infrastructure. These results suggest that, for large digital libraries and archives, retrieval accountability and corpus-aware agent design can provide a practical alternative to heavier training-based or graph-construction approaches.

As agent systems become more widely used, multiple agent sessions increasingly run alongside pre-existing user tasks in the same environment, sharing resources with limited capacity or mutually exclusive states. This creates a safety risk: when granted sufficient privileges, an agent may resolve a resource conflict by terminating or otherwise disrupting an existing task rather than reporting it. In this work, we identify and formalize this failure mode, which we term destructive resource preemption: obtaining the resources required for a requested task by terminating, overwriting, evicting, or degrading an incumbent task. To systematically study this risk, we introduce ClashBench, an executable benchmark comprising 268 validated conflict cases across 55 resource types, and evaluate 17 models through Codex, Claude Code, and OpenCode. We observe destructive preemption in 44.5% of trajectories, where the agent completes the requested task while causing the incumbent task to fail its health check. We also show that prompt-based safeguards are insufficient: an instruction to avoid affecting existing tasks reduces but does not eliminate preemption, while an instruction explicitly authorizing the agent to stop local processes increases it. More concerningly, in 31.9% of successful destructive-preemption cases, the final response mentions neither the resource conflict nor the action taken to resolve it, raising concerns about possible concealment. These findings establish destructive resource preemption as a broad safety risk in privileged agent systems and motivate stronger privilege controls, task isolation, and conflict-aware safeguards.

Gaussian Processes (GPs) are widely used as surrogates for black-box functions in sequential decision-making problems such as Bayesian optimization (BO), level set estimation (LSE), and Bayesian active learning (BAL). GP performance critically depends on kernels, and standard kernels can lead to suboptimal decisions under misspecification. To address this, we introduce HACK GPs (Hedge Adaptive Cumulative Kernels), a method that views kernel selection as an online learning with expert advice problem. HACK treats each candidate kernel as a GP "expert" and updates a distribution over experts online using AdaHedge, based on a loss received as a proxy for their ability to fit the function and align with the task objective. We provide two variants of HACK: (i) Mixture of Gaussians (MoG) and (ii) categorical sampling. We establish general guarantees showing that, under a loss-gap condition, the weight concentrates on the best kernel and the resulting acquisition function is close to that of the best expert. Empirically, we observe robust performance across BO, LSE, and BAL compared to standard kernels such as Squared Exponential and Matern-5/2, as well as simple ensemble baselines.

Interpreter agents built on machine translation (MT) increasingly mediate live conversation between people who do not share a language, yet we still evaluate them with metrics built for isolated sentences, which measure fidelity rather than whether communication succeeds. We introduce a reusable three-layer checklist-and-judge framework that evaluates interpreter-mediated conversation across semantic, pragmatic, and cultural-social dimensions, covering the naturalness, intent, and social appropriateness that fidelity metrics leave unmeasured. It runs in both single-turn and interactive multi-turn settings, where simulated users reply to translated messages as the conversation unfolds and each turn is scored alongside the conversation as a whole. We extensively validate it through controlled perturbations, cross-judge comparisons, and human annotations. Our main single-turn benchmark evaluates 10 interpreter setups across Arabic, Bengali, Indonesian, and Korean from 5,624 OpenSubtitles-derived scenarios spanning 12 translation directions, and our multi-turn study covers all 6 language pairs in scripted and live modes. Results show a consistent decline from semantic to pragmatic and cultural-social success, while conventional MT metrics overlook failures among stronger interpreters, and prompt ablations show that scenario context, structured instructions, and cultural context improve communicative success, although gains vary across setups. Our work thus provides an evaluation framework and benchmark for interpreter agents in conversation, and highlights the importance of communicative success alongside existing translation metrics.

Characterizing LLM reasoning remains an open challenge, as many existing benchmarks isolate specific reasoning skills, rely on external knowledge, or are costly to extend. We introduce PetriBench, a compact, fully self-contained, and scalable benchmark for evaluating LLM reasoning over dynamic state spaces using Petri nets, a mature formalism for modeling real-world concurrent and distributed systems. PetriBench organizes reasoning into four task families varying by scope and temporal horizon, with Easy, Medium, and Hard levels generated by increasing structural complexity and evaluated against exact ground truth. Across a diverse set of proprietary and open-weight models, accuracy decreases consistently with difficulty, while harder instances expose increasingly distinct task-specific capability profiles. Additional analyses show that test-time compute improves performance but interacts differently with different reasoning tasks, and that procedural generation yields smooth scaling with structural complexity. Together, these results show that PetriBench provides a unified and extensible setting for probing the strengths, limits, and scaling behavior of LLM reasoning.

Embodied visual perception relies on temporally coherent visual experience accumulated through continuous engagement with the environment. However, collecting large-scale egocentric binocular observations together with dense annotations remains costly and difficult. Moreover, visual experience is shaped not only by the environment but also by the embodiment of the observer, including viewing height, field of view, binocular geometry, and motion through the scene. To address these challenges, we present BinoGen, an automated framework for generating large-scale, embodiment-aware egocentric binocular visual experiences in indoor environments. BinoGen jointly models environmental and observer variation through generative scene synthesis, probabilistic object instantiation, appearance randomization, stochastic trajectory generation, and configurable binocular camera setups. The framework produces synchronized binocular videos together with dense multimodal supervision, including depth maps, optical flow, surface normals, semantic maps, object coordinates, and camera poses. Using BinoGen, we construct a dataset comprising more than 20 million annotated images for supervised learning. We demonstrate two complementary utilities of BinoGen. First, incorporating BinoGen data consistently improves real-world visual perception, including depth estimation, object detection, and video object tracking. Second, paired human-inspired and mouse-inspired observations from the same environments enable controlled investigation of how observer embodiment affects perceptual learning. Embodiment-specific adaptation substantially improves performance, while joint training enables a single model to perform competitively across both embodiments. Together, these results demonstrate that large-scale, controllable visual experience can improve embodied perception...

The KV cache has become a major bottleneck in deploying LLMs, as its memory footprint grows linearly with sequence length and batch size, imposing substantial pressure on both memory capacity and bandwidth. Among various KV cache compression techniques, quantization is particularly attractive due to its effectiveness and ease of deployment. However, most existing methods rely on fixed-width quantization, where a $b$ bit representation is inherently limited to $2^b$ quantization levels. As the bit width decreases, the number of available levels shrinks exponentially, leading to severe information loss and rapid performance degradation. We further observe that fixed-width quantization fails to exploit the highly non-uniform distribution of KV cache. After rotation and normalization, KV values approximately follow a normal distribution, with most values concentrated near the center and only a small fraction appearing in the tails. Nevertheless, fixed-width coding allocates the same number of bits to frequent and rare symbols. Entropy coding naturally exploits such non-uniformity by assigning shorter codewords to frequent symbols and longer ones to rare symbols, substantially reducing the average number of bits required for representation. However, its variable-length output is not suited to highly parallel attention kernels, where efficient dequantization and computation rely on regular memory layouts and fixed-stride accesses. To bridge this gap, we propose \textbf{D-Quant}, a flexible KV cache quantization framework that introduces a \textbf{drift} mechanism to convert entropy-coded representations of each token into fixed-size bitstreams, enabling regular memory access and parallel dequantization within attention kernels.

Question answering has advanced rapidly with large language models, but predominantly for high-resource languages, in both text and spoken settings. Spoken question answering (SQA) benchmark for Telugu remains unexplored, and the reliability of automatic evaluation in this setting remains unquantified. We introduce VākQA, a Telugu SQA benchmark of 2,001 factoid question-answer pairs across six domains, with 2.53 hours of speech audio, bilingual transcriptions, and human-verified reference answers. We first validate evaluation methods against human judgements: Gemini-as-a-judge best approximates human ratings but is non-uniformly strict, while open-weight judges systematically penalize correct Telugu answers that differ in surface form from the reference. Using this validated setup, we benchmark proprietary and open-weight models across input modality, language, and domain. We observe that Telugu phrasing retains cultural specificity that is lost in translation, speech input introduces phonetic confusions that alter question meaning, and cascaded ASR-MT errors compound progressively. VākQA is publicly released.

Multimodal reasoning requires models to draw on information from multiple modalities throughout the reasoning process. Yet existing methods often concatenate modality-specific thought tokens in a single sequence, leaving the model to bridge representational differences as it reasons across modalities. We introduce Uni-LaDiR (Unified Latent Diffusion Reasoner), a framework that brings these thoughts into a shared latent space for reasoning. A unified encoder maps teacher reasoning steps from different modalities into shared thought tokens, trained to preserve the information needed for later reasoning steps and the final answer or action. Because the same context can support multiple valid next steps, we use diffusion to predict the next block of thought tokens from the input and preceding blocks. Jointly training the encoder and diffusion reasoner with shared model weights encourages thought tokens to be both useful for the task and predictable from the available context. At inference, the model generates these tokens without teacher observations. Across eleven vision-language model (VLM) benchmarks and two vision-language-action (VLA) suites, Uni-LaDiR achieves relative gains over the strongest evaluated baselines of 7.3% on visual reasoning tasks and 6.1% on robot manipulation tasks.

Efficient long-term conversational memory requires retrieving sufficient evidence without indiscriminately expanding the context presented to the language model. This is challenging because relevant evidence may be distributed across multiple sessions, while compression may discard details needed for answering. Different queries therefore require different forms of memory access. To capture these demands, we formulate memory access along two dimensions: discovery breadth, which controls how broadly evidence is searched, and reading fidelity, which controls whether evidence is read in compact form or recovered from the original conversation. Based on this formulation, we introduce JustMem, which stores conversation history as compact atomic memories and adapts memory access along these two dimensions to each query. Specifically, LOOKUP handles local evidence, COMPOSE broadens discovery for distributed evidence, and REPLAY increases reading fidelity for fidelity-sensitive evidence. On LoCoMo and LongMemEval-S, JustMem achieves the highest mean accuracy and retrieval recall among the compared memory systems while using substantially fewer generative-model tokens for memory construction and inference.

Floor-plan-based indoor visual localization enables infrastructure-free positioning, but most methods are developed and evaluated in small residential environments unlike the large public buildings of real deployment. We introduce SlugTrails, a floor plan localization benchmark for large indoor spaces under realistic egocentric sensing: $30$ Hz Aria glasses recordings across three campus buildings and six floors ($22089$ m$^2$ of floor plan outline), CAD-derived floor plans with semantic classes and circulation space masks, and trajectories aligned into the floor plan frame using laser-surveyed anchors. One protocol covers three practical ways of gathering geometry under a limited field of view -- a single walking frame, a stationary multi-view sweep, and a walking stream with odometry -- so methods designed for different regimes are compared on the same buildings and ground truth. Evaluating five representative geometric and learned systems under their native sensing configurations, we find that stock checkpoints (official released weights) are near zero on SlugTrails (at most $0.004$ R@1m30$^{\circ}$ on walking single frames), while fine-tuning on SlugTrails improves every trainable family on all three tasks (e.g., F$^3$Loc $0.0 \rightarrow 0.141$ single-frame and $0.03 \rightarrow 0.66$ sequential), with gains compounding as observations accumulate. The same fine-tuned weights also improve cross-dataset generalization on LaMAR with no LaMAR training (sequential R@1m $0.048 \rightarrow 0.143$ for F$^3$Loc and $0.063 \rightarrow 0.127$ for UnLoc), whereas train-from-scratch on SlugTrails alone stays far below fine-tuning from stock weights -- evidence that floor plan localization is currently limited by indoor data rather than by architecture. We release the dataset, protocols, and tools at https://github.com/Head-inthe-Cloud/SlugTrails.

Autoregressive language models (ARMs) are constrained by sequential, left-to-right generation, while masked diffusion models (MDMs) enable parallel decoding but suffer from high computational overhead due to the inability to reuse Key-Value (KV) cache and from incoherent generation arising from learning dependencies over an intractable space of token combinations. We introduce Zarya, a family of hybrid language models that jointly optimizes an autoregressive (AR) objective and a masked-diffusion objective within a single architecture. Zarya structures training data into variable-size slots and employs a curriculum that gradually increases slot granularity, enabling a smooth transition from fine-grained AR learning to coarse-grained diffusion learning. At inference, Zarya provides two distinct decoding paradigms through a unified interface: (i) MDM sampling with first-hitting denoising, and (ii) slotted speculative decoding that interleaves inter-slot diffusion-based selection with intra-slot autoregressive infilling, achieving full KV cache reuse. The training and inference regimes are fully decoupled, allowing a model trained with any configuration to be deployed in either mode. Extensive configurability --- including grouped noise patterns (Prefix Completion, Fill-In-the-Prefix, Fill-In-the-Middle), ordered sampling schedules, and noise-level permutation strategies --- enables flexible research exploration. We release Zarya models publicly in sizes 0.6B, 1.7B, and 4B, demonstrating performance on standard benchmarks while offering a principled integration of autoregressive and diffusion paradigms.

Joint learning across heterogeneous tasks is often treated as task coupling through feature sharing, distillation, or auxiliary supervision. However, in cross-task learning, mismatched representational and supervisory granularities make such coupling prone to interference, teacher bias, or unidirectional collapse. We argue that cross-granularity learning is fundamentally a problem of hierarchical interaction regulation rather than simple task coupling. This issue is particularly evident in UAV perception, where visual shifts and detection--segmentation objectives naturally form coarse- and fine-grained knowledge sources. To systematically study this problem, we introduce CrossUAV, a UAV benchmark for joint object detection and instance segmentation that provides a unified evaluation platform for cross-granularity task collaboration. To address these challenges, we propose Cross-Granularity Socialized Collaboration (CGSC), a progressive and adaptive framework that regulates when, where, and how tasks exchange information across network hierarchies. CGSC progressively activates cross-task interactions and adaptively adjusts the strength according to task contribution, suppressing harmful interference while exploiting complementary coarse- and fine-grained structures. Extensive experiments demonstrate consistent improvements on both tasks, validating hierarchical dynamic interaction as an effective mechanism for cross-granularity collaboration.

Navigation world models plan by foresight, predicting the future that each candidate action sequence produces and selecting the best, rather than mapping observations to actions directly. Unlike urban settings where a predicted scene is a sufficient proxy, off-road navigation hinges on the robot--terrain interaction, so the prediction must cover not only what the camera will see but what the robot will feel. However, existing scene-focused models do not predict how much the robot will slip, tilt or shake along a planned trajectory. Proprioception captures these dynamics directly and, when used as input, improves the prediction of the physical future. We present Feel-WM, the first off-road navigation world model that conditions on proprioception and predicts what the robot will feel alongside what the camera will see. The physical future takes the form of a future proprioceptive state and a failure risk, both learned from the robot's own experience without human labels. The planner rolls out the physical future alongside the scene and weighs the predicted failure risk against goal similarity in a separable score. Experiments on real off-road data and in simulation demonstrate that Feel-WM outperforms visual-only navigation world models in open-loop planning and closed-loop rough-terrain navigation across wheeled and legged platforms. Deployed on a Husky on mountain trails, Feel-WM plans onboard, predicts rough ground ahead and steers around it, completing courses that an end-to-end policy fails.

MXenes have attracted considerable interest for integrated nonlinear photonics owing to their broadband optical response and chemical tunability, yet investigations have focused predominantly on Ti-based compositions, leaving the nonlinear potential of double-transition-metal MXenes largely unexplored. Here we quantify the nonlinear optical response of the double-transition-metal MXene \ce{Mo2TiC2T_x} and demonstrate its functionality as a waveguide-integrated nonlinear activation element for neuromorphic photonics. Z-scan measurements at \SI{800}{\nano\metre} reveal strong saturable absorption in films of two different thicknesses, yielding effective nonlinear absorption coefficients ($β_\mathrm{eff}$) between approximately $-2.69\times10^{3}$ and $-0.88\times10^{3}$ \si{\centi\metre\per\giga\watt}, with the nonlinear response decreasing at higher excitation intensities. The extracted nonlinear absorption is approximately one order of magnitude larger than that reported for \ce{Ti3C2T_x} MXene under comparable conditions. We integrate an ultrathin \ce{Mo2TiC2T_x} MXene layer onto a silicon rib waveguide to realize a compact nonlinear optical activation function, which, when implemented in a neural-network emulator, achieves $98.39\%$ classification accuracy on the MNIST benchmark. This work expands the MXene material platform beyond Ti-based compositions and establishes double-transition-metal MXenes as promising candidates for high-performance integrated nonlinear photonic and neuromorphic computing technologies.

As generalist robot policies gain vision and language from web-scale pretraining, demonstrations remain costly to collect and tied to the robot that recorded them. Latent action models (LAMs) address both by learning latent actions from action-free videos that can be shared across embodiments, however, in practice, LAMs are sensitive to background visual noise, and the same motion from two different robots may be encoded with different latents. One solution to the background visual noise is to add an auxiliary loss predicting the robot action from the latent action, further associating the latent action space to the embodiment specific robot action space. We study a different use of the same labels, through action-similarity supervision. The similarity between any two latent actions is trained to match the similarity of the two ground-truth robot action sequences. The ground-truth actions are never predicted by the LAM, so the latent action does not need to encode embodiment specifics. We evaluate cross-embodiment transfer on RoboTwin 2.0 in a controlled setup, two bimanual robots demonstrate disjoint task sets, a policy is trained on all the demonstrations, and each robot is evaluated closed-loop on the tasks only the other demonstrated. With the policy architecture and its hyperparameters, the dataset, and the evaluation protocol fixed, predicting latent actions instead of ground-truth actions more than doubles cross-embodiment success. Given the same ground-truth actions, similarity supervision transfers better than an auxiliary loss that predicts the ground-truth action during the LAM training. Computing the similarities on end-effector motion rather than joint-space motion, and letting the loss compare latent actions across the two robots, gives the best approach of the study.

AI-generated RTL verification plans can satisfy a provider schema yet fail at the boundary to trusted execution. We present SecTB-RTL, an auditable framework covering 31 tasks and 124 authored hardware-security regressions. A deterministic non-AI baseline killed 36, 75, and 78 mutants at increasing resource limits. The first confirmatory run (C1-R2) failed before model execution because the provider rejected its response schema. After a schema-only repair made without viewing outcomes, a separately frozen follow-up run (C1-R3) completed 1,860 calls. The provider accepted 1,857 responses, but only nine passed the production semantic validator. The generation and execution rules did not match. We therefore preserve the run as an instrument-validation incident and report no prompt-effect estimate. This incident shows that provider or schema acceptance does not establish execution validity. Compilation and coverage are only diagnostics; the exact saved artifact must pass the full production path. A subsequent follow-up is excluded because it did not satisfy the preregistered evidence-completeness gate and is treated only as future work. We release the benchmark, failure-preserving contract, incident provenance, and governance controls needed to prevent infrastructure behavior from being misreported as model behavior.

High-quality 3D talking face datasets remain largely English- centric, and Korean 3D facial motion data are difficult to combine with standard English benchmarks because of differences in mesh topology, spatial scale, coordinate system, and temporal sampling. We present KoUniTalk, a lightweight articulation-centered Korean-English 3D talk- ing face benchmark that retargets VOCASET and the released Korean speech-based 3D talking face data to a shared mesh topology using de- formation transfer. Rather than proposing a new deformation-transfer algorithm or a full-head identity-preserving avatar dataset, KoUniTalk provides an identity-neutral canonical output space for controlled speech- driven facial articulation training and evaluation across English and Ko- rean. The unified template contains 1,176 vertices and focuses on the mouth and adjacent lower- and mid-face regions, reducing the output dimensionality from 15,069 and 72,147 dimensions to 3,528 dimensions, corresponding to 4.27-fold and 20.45-fold reductions compared with VO- CASET/FLAME and the original Korean mesh, respectively. To exam- ine whether retargeting preserves speech-relevant motion, we evaluate semantic mouth-landmark trajectories, including mouth opening, mouth width, aperture ratio, and mouth-opening dynamics. Since the official test set of the Korean dataset is not publicly released, we additionally define a subject-disjoint Korean benchmark split. The processed matched benchmark contains 22 speakers, 4,978 sequences, and 642,781 frames, enabling Korean-English cross-dataset evaluation of speech-driven 3D fa- cial animation models in a single compact articulation-template space. Source-reported inventory counts are listed separately from these pro- cessed counts

With the widespread industrial deployment of Large Language Models (LLMs), DeepSearch has emerged as the dominant paradigm for resolving complex user queries. It typically operates through an iterative closed-loop workflow consisting of planning and reflection, information retrieval, and answer generation. However, existing reward models (RMs) and evaluation benchmarks are primarily designed for static single-turn tasks, failing to capture the full-pipeline complexity of DeepSearch workflows. To address this limitation, we propose F2DR, a fine-grained full-pipeline DeepSearch reward framework. F2DR evaluates DeepSearch workflows across three dimensions: Content, Trajectory, and Answer, enabling comprehensive process-level assessment. We further construct DeepSearch RM-Bench, a dedicated benchmark for evaluating RMs in DeepSearch scenarios. Extensive experiments demonstrate that F2DR achieves significantly higher evaluation consistency than self-evaluation-based baselines, while DeepSearch RM-Bench exhibits strong discriminative capability across existing open-source RMs. We will publicly release the complete DeepSearch RM-Bench dataset soon.

Regularized self-play -- the family behind DeepNash's Stratego play -- drives a two-player zero-sum policy to a Nash equilibrium by best-responding to a slowly moving, entropy-regularized reference policy $ρ$. When the game has a polytope of value-equivalent equilibria, the regularizer silently breaks the tie: with a uniform reference it selects the maximum-entropy member, the I-projection of $ρ$ onto the Nash set. Can the reference be used to choose the equilibrium on purpose? On five exactly solvable games plus a 2-D polytope, with exact best responses and equivalence tests over independent seeds, anchoring the reference at a target member and refining steers self-play to that member with mean coordinate error 0.007 at median exploitability $5\times10^{-5}$, TOST-equivalent to the request within $\pm0.05$; the anchoring persists through refinement and follows the reference, not the initialization. Selection follows the reach-weighted I-projection (slope 0.969 [0.950, 0.987]). We report with equal emphasis where the story breaks: fixed off-manifold references cost 0.08-0.25 exploitability; stiff or flat families require a smaller mirror step, set by a pre-registered rule; boundary targets undershoot; curvature predicts where boundary saturation bites (rank correlation 0.90, p=0.037) while interior precision is curvature-independent. Table and MLP steering maps are equivalent within $\pm0.03$ at every target (30 seeds); matched control arms show attention's robust signature is excess seed variance, any systematic shift bounded at 0.018 and not significant. Against a best response the selection-robustness trade-off is degenerate: steering matters only against fixed, non-equilibrium opponents. The recipe -- anchor the reference at the desired member and refine -- reinterprets the KL anchor of RLHF-style RL as a selection knob, not only a stability leash.

Robotic garment unfolding is essential for downstream tasks, yet quasi-static methods require repeated actions, while existing dynamic approaches predominantly rely on bimanual flinging. We present RotateIt!, a single-arm framework that uses adaptive axial rotation for dynamic garment unfolding. To the best of our knowledge, it is the first unfolding framework to employ dynamic axial rotation as its primary manipulation primitive. From a randomly initialized tabletop configuration, the robot selects a rotation-effective grasp and rotates the lifted garment about an approximately fixed anchor, generating inertial tension that separates overlapping layers within a compact workspace. A grasp ranker selects the anchor, while an online residual policy adapts the rotation extent and speed, thereby determining the release timing. Across seen and unseen simulated garments and eight unseen real garments, RotateIt! improves success within three attempts by 44.0-61.0 percentage points over quasi-static pick-and-place. The simulation-trained policies transfer zero-shot to the real world, achieving 75.6% success, 41% higher first-attempt coverage, and 26% higher final coverage. The resulting states further enable autonomous robotic folding without manual rearrangement.

We propose SnapPhysics, a training-free framework that reconstructs 3D objects and estimates their physical properties such as mass, friction, and center of gravity from a single image. For physically coherent interactions in mixed reality (MR), such properties are as important as geometry. Prior approaches infer them by analyzing object dynamics in video, which is computationally costly, or by querying vision-language models (VLMs) on single images, which lacks geometric grounding and inter-object relationships. We address these limitations by combining instance-level 3D reconstruction and spatial alignment with a physics-aware scene graph that encodes these relationships and per-object metric geometry as structured context for VLM-based property reasoning. Experiments on 3D-FRONT show that SnapPhysics improves scene-level F-Score by 18.6% over the best learning-based method, and on real captured scenes with ground-truth mass, it reduces the mean absolute log difference error (mALDE) by up to 20.5% and improves log-scale correlation ($r^2_{\mathrm{ls}}$) by up to 19.6% over VLM-only estimation. SnapPhysics enables physically interactive MR experiences without manual parameter tuning. Project page: https://snapphysics-ismar2026.github.io/.

Predicting the future trajectories of surrounding vehicles in autonomous driving is important for collision risk assessment and safe ego-vehicle path planning. Conventional neural network-based trajectory predictors typically achieve strong prediction performance by exploiting agent history, dynamic scene graphs, and semantic maps. However, in specific motion regimes such as acceleration, deceleration, and turning, these predictors may fail to reflect physically feasible trajectories. To address this issue, this study proposes a framework that fuses the output of Trajectron++, a neural network-based trajectory predictor, with extended Kalman filter (EKF)-based multiple trajectory candidates at a late stage. On the nuScenes dataset, the proposed method reduces the average displacement error and final displacement error of the Trajectron++ robot baseline by 13.7% and 14.6%, respectively, without modifying the baseline architecture. These results indicate that EKF-based trajectory candidates can effectively complement neural trajectory prediction through learned fusion.

Grapheme-to-phoneme (G2P) conversion turns raw text into its phonemic form and is an essential part of both text-to-speech (TTS) and automatic speech recognition (ASR) systems. It is required to be fast, stable and context-aware. For unsegmented languages such as Japanese, G2P additionally couples word segmentation with highly context-dependent polyphone disambiguation, and the scarcity of accurately annotated data remains a bottleneck. In this paper, we present a context-aware neural G2P method that scores paths of a discriminative conditional random field (CRF) over a word lattice constructed from dictionaries. To tackle data scarcity, we utilize large language models (LLMs) to generate more than 2 million sentences. Experimental results demonstrate that our method strongly outperforms conventional morphological analyzer-based methods and neural sequence models. On the Joyo-Kanji-Yomi benchmark, our method reaches 99.62% target word reading accuracy, 0.32% target word phoneme error rate (PER) and 0.14% sentence PER.

Multi-agent heterogeneous air-ground robot teams are attractive for open world search, with applications for reconnaissance, urban search and rescue missions (USAR), disaster response and recovery, and hazardous environments. These two platforms have different failure modes: aerial robots cover ground quickly but cannot resolve small or occluded targets from altitude, while ground robots can identify objects-of-interest, such as people or hazardous objects, at close range but cover less area. Existing language-tasked teams either have roles fixed prior, or have a language model assign them from hand-written capability tags, so the team is unable to know when within a mission an asset is no longer useful. We present HEROIC, a decentralized heterogeneous multi-agent open-vocabulary search coordination framework that requires agents to communicate in natural language only. HEROIC's initial agent role assignment is derived from sensor properties and a scale law to determine whether targets can be detected with a high confidence. From the mission's natural language prompt alone, this law assigns aerial flight altitudes and sweep spacing. When this calculated height falls below the altitude for safe flight, aerial agents re-task themselves from searcher to aerial triage, escort, and route guide for ground agents. Both robots maintain an evidential belief over the search area (bearing rays for positive evidence, a log-odds posterior for negative evidence) and gate any arrival on close-range verification. In full-stack experiments, HEROIC reaches the target 84% of the time across all 6 scenes, compares to 35-54% for vision-language frontier baselines, frontier-based search, lawnmower, and random-walk running the same perception, all while being 2-4x sooner to arrive at the target.

Physical teleoperation integrates human cognitive flexibility with robotic precision, yet demanding manipulation tasks frequently induce severe cognitive workload, acute frustration, and execution breakdown. Conventional shared autonomy paradigms rely primarily on task-based rules, such as spatial error boundaries, which disregard the operator's transient affective state and risk misaligned control interventions. To address this limitation, we propose an affect-aware shared autonomy teleoperation framework that dynamically modulates robotic assistance based on real-time operator state estimation. The system estimates operator affective states from synchronized facial video, cardiac signals, and bilateral arm kinematics, outputting a seven-state affective distribution and a three-category operational abstraction (neutral, productive, adverse). Affect-aware assistance is selectively triggered when the user is detected in a continuous adverse state, preserving task-positive engagement without unnecessary disruption. The empirical user study ($N = 30$) confirms that the proposed affective assistance increases the productive states by up to 39.7% without compromising user agency. The collected dataset represents the first multimodal dataset that provides continuous visual, physiological, and operator's bilateral motion tracking of temporal affective state shifts during bimanual teleoperation. Our multimodal fusion model outperforms zero-shot baselines (Qwen, MiniCPM-V) in tracking temporal state dynamics. This real-world deployment offers a new human-centric framework that integrates visual, physiological, and motion tracking for physical human-robot interaction.

Executable environments enable LLM agents to learn from the consequences of their actions. For embodied agents, those consequences extend beyond whether the current task succeeds: completing a delivery can consume the time, energy, or money needed for later work. Learning to plan therefore requires environments that preserve these dependencies and turn them into feedback across a complete trajectory. We introduce DeliveryGym, a 3D environment for evaluating and training agents on continuous courier shifts. It couples multimodal tool interaction with persistent world dynamics and computes trajectory rewards from simulator events, making the costs of an agent's decisions available for reinforcement learning (RL). The environment also adapts future training shifts to the policy's observed weaknesses while keeping evaluation fixed. Across six models and 13 city maps, evaluation exposes a gap between reliably executing assigned deliveries and choosing and sequencing work over a shift. On the fixed test suite, RL improves Qwen3-VL-4B's net income by 54.3%, showing that learning from complete shifts improves performance under these coupled constraints. Adapting the training environment improves test income by 16.5% over uniform sampling at the same rollout budget, indicating that which situations an agent practices also matters. DeliveryGym provides an executable setting for studying how agents learn to coordinate deliveries and preserve resources for later orders within an episode.

Robotic manipulation under partial observability requires spatial information that extends beyond the current view. Geometry-aware RGB features describe visible structure, but previously observed regions may disappear as the robot or scene moves. Maintaining a useful scene representation therefore requires retaining observation history while inferring missing content without losing its connection to visible evidence. We introduce LIFD (Look, Imagine, Focus, and Do), a framework for persistent, 3D-aware scene memory. LIFD learns a scene-token representation from multi-view agreement and completes it from a single RGB view and recurrent memory. A rectified-flow model generates the tokens while Anchor-Guided Cross-Attention conditions completion on current geometric features. Compact slot features connect this representation to a manipulation policy. Multi-view and geometric supervision are used during representation learning; deployment requires one RGB camera, proprioception, and a task instruction. LIFD (Staged) reaches 91.6% average success on LIBERO and 79.8% on MetaWorld, improving LIBERO average success by 3.1 percentage points over Joint training. On four UR5e task families with ten demonstrations per family, it achieves 56.0% mean success, compared with 40.5% for OpenVLA-7B.

Recent advances in artificial intelligence (AI) are reshaping smart glasses from egocentric capture and display devices into platforms for wearable intelligence. Smart glasses increasingly serve as wearable AI systems that connect first-person observation with real-time assistance under strict form-factor constraints. We frame this transition through the lens of \emph{AI smart glasses} and define them as a system-level concept in which egocentric sensing, resource-aware computing, intelligent reasoning, multimodal interaction, and real-world application constraints are co-designed for personalized assistance in the physical world. To systematically study this perspective, we organize the survey around four connected dimensions. First, we examine the hardware foundation that bounds sensing, computation, feedback delivery, and sustained deployment. Second, we study wearable intelligence, where egocentric signals are transformed into perceptual, contextual, and agentic capabilities. Third, we discuss interaction design, through which users request, receive, correct, and regulate assistance during ongoing activity. Fourth, we analyze application scenarios across healthcare, accessibility, situated learning, daily life assistance, cultural tourism, and industrial support, showing how domain requirements reshape system design and evaluation. We further identify five cross-cutting research challenges for future AI smart glasses: next-generation hardware, trustworthy egocentric intelligence, lifelong personalized memory, proactive intelligence, and embodied foundation models. By centering smart glasses as wearable-intelligence platforms, this survey provides a unified framework for organizing technologies, applications, and open challenges in this emerging area.

The inverse problem of reconstructing optical properties, specifically absorption and scattering coefficients, in layered biological media from time-domain reflectance measurements remains a significant challenge for traditional analytical models. Inverse solvers based on the diffusion equation often struggle with structural heterogeneity, frequently yielding poor accuracy for superficial absorption and deep-layers scattering. In this work, we propose a machine learning framework as an alternative approach to reconstruct the optical properties of a bilayered medium, benchmarking its efficiency and accuracy against model-based algorithms. To overcome the intrinsic approximations of diffusion theory and inverse reconstruction, we generated a robust synthetic dataset of forward DTOF using exact Monte Carlo simulations at multiple source-detector distances. A machine learning pipeline was then trained on this dataset and validated against state-of-the-art model-based reconstruction methods. Besides the significant reconstruction speed-up, the machine learning approach achieves higher accuracy than model-based inverse solvers, further providing an estimate of the parameter space dimensionality without requiring any a priori information about the number of layers in the investigated geometry. Further enhancements in the reconstruction accuracy can be expected in future extensions of this work, by training the pipeline over multiple DTOF curves from the same medium, in a joint multi-distance reconstruction approach.

Hateful memes spread abusive content through implicit interactions between images and text, posing serious threats to the safety of online communities. In recent years, multimodal large language models have been widely used for hateful meme detection and are increasingly adopted to generate explainable detection results. However, we find that existing explain-then-detect methods often couple explanation generation and label prediction within the same training process. This coupling causes interference between task objectives, leading to limited detection performance and even worse results than simple SFT baselines. To address these challenges, we propose ProKDA, a progressive knowledge-to-decision alignment method for explainable hateful meme detection. Inspired by the human annotation training process, ProKDA first uses an agentic background knowledge construction pipeline to obtain external knowledge related to meme understanding. It then adopts a three-stage training strategy that sequentially performs background knowledge learning, hatefulness detection learning, and hatefulness boundary alignment. Unlike prior explain-then-detect methods that jointly optimize both tasks, ProKDA focuses on a single training objective at each stage. This design reduces interference between the two tasks and progressively transforms background knowledge into robust detection decisions. Experiments on three public hateful meme benchmarks show that ProKDA achieves state-of-the-art detection performance and provides accurate, explainable, and evidence-supported decisions for hateful meme moderation. Project page: https://meizhiyuan88666.github.io/prokda.

All-atom structure predictors model diverse molecular interactions, but using their learned structural priors for binder design remains challenging. Here we present TorchCraft, a unified binder-design framework that optimizes sequence logits through a frozen all-atom predictor. Implemented in TorchFold, TorchCraft combines confidence, contact, geometric, and sequence-prior objectives within a shared optimization procedure for minibinders, framework-conditioned VHHs, cyclic peptides, and ligand-binding proteins. Using pretrained AlphaFold 3 weights, TorchCraft generated representative minibinders and VHHs with experimentally measured binding across four targets in each format, without post hoc sequence redesign. Computational benchmarks further demonstrated the framework's applicability to cyclic peptides and ligand-conditioned pocket design. TorchCraft extends predictor inversion to multiple binder formats and molecular contexts, providing a common framework for reusing all-atom structural priors in design.

True machine intelligence requires transcending passive pixel registration to master top-down functional reasoning over absent information via visual negation understanding. However, unconstrained visual negation paradigms remain overly open-ended, and pervasive affirmation bias causes both existing Multi-Modal Large Language Models (MLLMs) and evaluation metrics to fail under negative semantics. To solve these intertwined challenges systematically, we first anchor the boundaries of negation reasoning within specific cognitive goals. Specifically, by focusing on safety as a highly pragmatic and critical cognitive dimension, we define the task of \textbf{S}cene \textbf{N}egation \textbf{U}nderstanding under \textbf{S}afety Cognition (\textbf{SNUS}). Under this framework, we construct a high-fidelity negative caption dataset mapping dense assertions of localized hazards. Concurrently, we propose the Cognitive Expected Scene Graph (CESG) Score, a structure-grounded, polarity-aware evaluation metric. Extensive experiments demonstrate that while current models struggle on the task, traditional metrics completely collapse under semantic reversals. Conversely, our framework delivers a solid benchmark for SNUS, providing a rigorous foundation to advance risk-aware situational comprehension and counterfactual cognition.

Accurate classification of brain disorders from neuroimaging data remains challenging because of substantial inter-subject heterogeneity and the complex multi-scale patterns present in functional connectivity and morphological representations. To address these challenges, we propose HyperAMS-Net, a deep learning framework for brain disorder classification using neuroimaging representations derived from resting-state functional MRI or structural MRI. HyperAMS-Net integrates adaptive multi-scale convolution, hypergraph attention, spatial-channel attention, and adaptive feature fusion. Specifically, adaptive multi-scale convolution learns data-driven weights over multiple receptive fields to capture complementary patterns at different scales. Hypergraph attention models higher-order dependencies among learned feature representations through node--hyperedge--node message passing, while spatial-channel attention enhances discriminative feature learning. Adaptive feature fusion further aggregates complementary information across parallel network branches. HyperAMS-Net is evaluated on three benchmark datasets spanning distinct brain disorders: ABIDE for autism spectrum disorder, REST-meta-MDD for major depressive disorder, and ADNI for Alzheimer's disease, using 5-fold stratified cross-validation. HyperAMS-Net achieves state-of-the-art performance across all evaluated datasets, attaining the highest accuracy and AUC among the compared methods. Ablation studies further demonstrate the contribution of each proposed component, with the largest performance degradation observed when hypergraph attention is removed.

LLM agents have recently shown promise in automating machine learning engineering by editing model and training code under execution feedback. Data, however, remains largely outside this agentic optimisation loop. We frame pre-training data selection as heuristic engineering over per-document features, i.e., lexical statistics, categorical labels, and perplexity. We introduce AutoData, an agent that searches directly over executable selection algorithms. Unlike prior data mixture methods that optimise weights over a fixed set of domains, AutoData searches a richer program space of scoring, stratification, and stochastic selection rules, discovering feature interactions automatically by iteratively refining algorithms with validation feedback from a proxy model. Within an overnight search, AutoData discovers a selection algorithm that outperforms existing human-designed curation pipelines. Despite being searched only on this small proxy, the discovered recipe transfers to larger scales and improves the downstream metric CORE. These results suggest that data engineering can be treated as an agentic machine learning problem, extending autonomous research from model and training-code optimization to the data.

Fine-grained visual perception in MLLMs is commonly improved by raising the resolution, but the added visual tokens inflate vision-encoding and language-model prefilling costs. We show that the two operations underlying fine-grained perception, localizing the region of interest (RoI) and recognizing its content, have different resolution requirements. In a controlled diagnostic, localization tolerates roughly 3 to 4 times stronger token compression than recognition, which motivates localizing from a coarse view and concentrating resolution on the selected evidence. Decoding coordinates with the MLLM can be trained end-to-end from answers, but costs a full model pass per query and depends on grounding ability. A lightweight proposal network distilled from the model's attention is fast, but inherits the noise of its attention targets. The RoI from the proposal network reaches the answer through a discrete region choice, so its faithfulness to the answer cannot supervise the network. We therefore optimize the proposal network with region-level reinforcement learning, which we call Vision-RL2. It treats coherent regions as actions, and a frozen MLLM reader scores each one by how its removal changes the answer likelihood. Complementary subtractive and additive objectives suppress distracting proposals and recover missing evidence, updating only the predictor without region annotations, response sampling, or reasoning trajectories. The refined proposal further enables a sparse encoding that magnifies evidence and excludes background tokens. Across six fine-grained benchmarks and four MLLM backbones, Vision-RL2 improves accuracy over the base model at every token budget and surpasses its largest-budget accuracy with about 4 times fewer visual tokens. Code is available at https://github.com/YuHengsss/VisionRL2 .

Autonomous Underwater Vehicles (AUVs) navigating without GPS typically fuse inertial measurements with acoustic Doppler Velocity Log (DVL) velocities and pressure-derived depth. Posing the navigation state on a Lie group improves accuracy and consistency. However, state-of-the-art filters based on the Invariant Extended Kalman Filter (IEKF) append the Inertial Measurement Unit (IMU) biases as a Euclidean extension, which breaks the group-affine structure required for exact log-linear error dynamics, causing the reported covariance to degrade alongside the estimate. We apply the Tangent-Group (TG) symmetry, which carries the biases within the geometry of the state space, to derive an Equivariant Filter (EqF) for this system, leaving zero linearization error in the navigation states and second-order error only in the biases. We develop an equivariant output model for the DVL, whose update incurs only third-order linearization error, together with a direct pressure output. Monte Carlo simulations benchmark the TG-EqF against a Two-Frame-Group IEKF and a Multiplicative EKF. The TG-EqF reduces error by 18--25\% against both alternatives in each of attitude, velocity, and position. The main benefit is in the covariance it estimates: its Average Normalized Estimation Error Squared (ANEES) stays closer to its nominal value of one than that of the others. Offline analysis on AUV field data corroborates the findings of the simulations, demonstrating reduced position drift.

Autoregressive (AR) video generation degrades over long horizons due to an overlooked train-inference discrepancy we term KV eviction mismatch: models train on short clips where all context frames reside in the KV cache, but at inference, memory constraints force distant frames to be evicted from the KV cache - removing context the model was conditioned on. Rather than simulating eviction via context truncation - which discards temporal information the model still needs and degrades motion coherence - we keep the context but while progressively reducing the influence of distant frames, making their eventual eviction negligible. To guide this design, we introduce the positional response $R( Δ, \, t_{\text{denoise}})$, a perturbation-based sensitivity measure revealing that context influence decays steeply with temporal distance and varies systematically across denoising steps. Motivated by this analysis, we propose Recency Forcing, which applies a non-positive, timestep-dependent bias, termed Temporal Response Bias (TRB), on pre-softmax attention logits derived directly from $R$, closing the train-inference gap without modifying context length or training objectives. We further introduce Biased Attention Reparameterization (BAR), an exact reformulation that moves the bias outside the softmax, making TRB a standard FlashAttention call at zero overhead. Recency Forcing operates in both training-free mode and training-based mode. Experiments on VBench and VBench-Long demonstrate state-of-the-art long-horizon generation quality at no additional inference cost.

Clinical coding agents repeatedly encounter the same failure modes, including unsupported codes, missed documented conditions, specificity errors, and procedure-coding convention mismatches. We introduce Learn-Then-Act, an inference-time adaptation framework that converts errors from a small labeled LEARN batch into a structured Mistake Knowledge Database (MistakeKDB). False-negative lessons are routed to a recall-oriented Coder, while false-positive lessons are routed to a precision-oriented Judge. We instantiate the framework in LearnActCoder, a Coder-Judge clinical coding pipeline with lookup-table grounding where available. On 150 matched MIMIC-III notes, structured MistakeKDB improves CPT F1 by 5.9 percentage points, while raw-example and reflection-style memories remain near the no-memory baseline; the ICD-9 improvement is not significant. On a matched MIMIC-IV cohort, memory shifts ICD-10 coding toward higher precision at a recall cost, leaving F1 statistically unchanged. Applying the same memory to 1,000 held-out MIMIC-III notes maintains a stable ICD operating point, providing scale/stability evidence. Overall, the results are consistent with structured, feedback-derived error memory being useful for adapting clinical coding behavior across cases without weight updates or changes to the underlying workflow. Absolute CPT/HCPCS performance remains low, and the system is evaluated retrospectively rather than in clinical deployment.

Large Language Models (LLMs) have achieved remarkable reasoning capabilities by utilizing chain-of-thought (CoT) as a scratchpad for intermediate stages of thinking. However, CoT techniques require explicit supervision on thinking tokens, which requires rich, task-specific data. In this work, we propose Abstract Token Curriculum (ATC), a novel curriculum learning framework that elicits effective continuous intermediate representations without direct supervision or manual scratchpad design. ATC gradually increases problem complexity through a sequence of distributions, training the model to develop internal abstract ``thoughts'' in the continuous representation space. This paper provides both theoretical and experimental evidence for the benefits of ATC and its advantages over previous methods for training continuous thoughts. Theoretically, we show that for learning parity functions with single-layer softmax attention using ATC, attention naturally focuses on the CoT tokens in the context that provide the ``easiest path'' to predicting the next token. Experimentally, we show ATC's effectiveness on graph reachability and arithmetic learning tasks.

Pre-trained 3D vision models have substantially advanced point cloud analysis, yet adapting them to downstream tasks via full fine-tuning is computationally expensive and storage-intensive. Parameter-Efficient Fine-Tuning (PEFT) offers a promising alternative by reducing both adaptation cost and storage burden. However, existing prompting-based approaches ignore the intrinsic geometric structures of point clouds, thereby limiting their adaptation capability. This limitation stems from their inability to encode both fine-grained geometric cues and coarse-grained structural semantics, as well as failing to propagate such information effectively through the model hierarchy. To address these challenges, we propose GAPrompt++, a multi-granular geometry-aware prompting method that provides richer geometric guidance for efficient 3D task adaptation. Specifically, we introduce a Point Shift Prompter that extracts multi-granular geometric features across different scales, enabling instance-specific geometric adjustments during adaptation. Next, a Keypoint Prompter adaptively generates point-level prompts to highlight local geometric saliency and fine-grained structural details. Furthermore, a Prompt Propagation mechanism injects these multi-granular geometric cues throughout the feature extraction hierarchy, strengthening the ability to capture essential geometric characteristics. Extensive experiments show that GAPrompt++ achieves state-of-the-art performance among prompting-based PEFT methods and even surpasses full fine-tuning across diverse benchmarks, while requiring less than 2\% trainable parameters. In addition, to address the saturation of existing evaluation datasets, we construct two more challenging benchmarks derived from 3D Gaussian Splatting and Multi-View Stereo reconstruction, offering diverse and realistic point cloud scenarios to promote future research.

Autonomous large language model (LLM) agents are moving rapidly into high-stakes domains, yet existing agentic-AI security studies remain largely domain-agnostic and overlook the distinctive, high-consequence attack surface such settings create. We examine this gap through financial trading agents, a representative case of high-stakes agentic security, where a single compromised agent has direct execution authority over real capital in an adversarial, reflexive market. To this end, we present FARSIGHT (Financial Agent Robustness and Security Investigation and Global Holistic Testing), a framework that performs scheme-level evaluation of financial LLM agents on two axes: robustness under market turbulence (including flash-crash-like scenarios), and security against three attack types: attacks on information sources, attacks on agents, and agent-as-attacker behaviors. Applying FARSIGHT to 15 representative academic schemes, we find that most overlook robustness and realistic adversarial threats: 80% fail at least one core robustness metric and 100% exhibit security vulnerabilities. These two failure modes are inseparable: a small misjudgment can cascade into a market-wide crash on its own, while an adversary can deliberately trigger the same collapse at minimal cost.

Daily locomotion encompasses diverse activities and frequent transitions between them, yet most exoskeleton controllers are designed for a single activity or a narrow set of related movements. Changes in activity therefore typically require explicit mode switching and separately tuned or retrained controllers. Simulation-based learning reduces the need for hardware-based tuning but generally retains this limitation. Here we present UniExo, a framework that first constructs a multi-skill musculoskeletal human policy and then jointly trains an exoskeleton control policy with it. Four single-skill imitation experts for walking, turning, running and backward walking are distilled into a single network structured by a skill latent and subsequently fine-tuned through reinforcement learning on transition sequences. The resultant unified human policy achieves a mean tracking success rate of 94.7% on unseen clips of the four skills and exhibits greater robustness to perturbations than its constituent experts. A single hip exoskeleton controller (UniExo) is initialized from hip moment prediction of the human policy and co-adapted with it through multi-agent reinforcement learning across the four skills. This co-adaptation shifts the timing of the assistance torque and raises the fraction of positive work delivered to the hip. When deployed on a custom hip exoskeleton, the controller generalizes across four treadmill speeds in six participants and assists one participant through a continuous route of all four skills and their transitions, without skill labels or explicit mode switching. UniExo thus provides a step towards replacing activity-specific controllers with unified, user-specific controllers that support diverse locomotor activities and the transitions between them.

We present LYRIC, a generative flow-matching controller for language-driven physics-based contact-rich interaction control, that enables simulated characters to perform contact-rich whole-body object interactions from a free-form language instruction and a sparse terminal object goal. To obtain reliable expert trajectories from imperfect motion-capture references, a single tracking policy is trained using geometry-conditioned interaction rewards and relaxed reference tracking near hand-object contact. To guide interaction progress without prescribing a full-body kinematic reference, we factorize the controller into a task-level planner that predicts short-horizon object and humanoid-root trajectories, and an action generator that resolves whole-body motion and contacts in closed loop. After behavior cloning, we freeze the planner and post-tune the action generator on policy using the planner's predictions as stable supervision for intermediate task progression. In a controlled OMOMO evaluation, our tracker achieves 64.3% success compared with 53.2% for an InterMimic reimplementation, while a unified policy achieves 76.5% on the full OMOMO dataset. On the held-out split, LYRIC achieves 90.3% task success, compared with 74.2% for the strongest matched kinematic-planner baseline, with better semantic alignment and motion quality. Without retraining, the controller also supports test-time object-waypoint guidance. Qualitative results further demonstrate robust, natural contact-rich interactions and zero-shot transfer to novel object shapes. The webpage is available at https://neu-vi.github.io/LYRIC/

The deployment of Vision-Language Models (VLMs) on edge devices is severely bottlenecked by memory bandwidth, necessitating aggressive sub-8-bit quantization. Since edge accelerators are strictly constrained by area and power, they require end-to-end quantized models. However, the extreme dynamic range gap between multi-modal tokens causes standard block formats to suffer "microscaling collapse," where a single massive outlier hijacks the shared exponent, underflowing surrounding elements and destroying attention maps. To break this bottleneck, we propose Micro-Inverted-Scaling (MiX), a novel format that mathematically inverts the microscaling paradigm: rather than grouping multiple mantissas under one shared exponent, MiX groups private, per-element exponents under a single shared mantissa. To handle asymmetric VLM outlier topologies, we introduce an adaptive dual-format (MiX-MX) inference framework. By algebraically factoring out the shared MiX mantissa, this framework maps to a custom accelerator, replacing multipliers with efficient shifters. Evaluated end-to-end on multiple VLMs, our 4.5-bit MiX formulation exhibits equivalent or superior accuracy on multi-modal benchmarks compared to NVFP4. Simultaneously, the MiX accelerator delivers a 25% improvement in area efficiency over the NVFP4 baseline and a 2.3-4.5x speedup with 1.4-2.9x energy reduction across models compared to the state-of-the-art accelerator Focus, proving the inverted-scaling datapath is physically superior for efficient VLM deployment.

Financial QA systems are typically improved before deployment through better retrieval, prompting, or agent coordination, leaving their reliability behavior fixed thereafter. In practice, new SEC-filing questions repeatedly expose heterogeneous errors in period, entity, evidence use, and calculation. Existing self-improvement methods can turn failures into new behaviors, but offer limited control over where a correction should apply or which previously correct answers it may break. We therefore frame post-deployment improvement as controlled behavioral maintenance: recurring failures should become scoped skill patches, and each patch should earn deployment with- out introducing regressions. We instantiate this view in FINSKILLOPS, a multi-agent system for SEC filing QA. FINSKILLOPS derives reusable skills from evidence-grounded, typed failure diagnoses and governs them through targeted validation, protected-case regression checks, negative controls, and versioned replacement or retirement. Across six financial QA benchmarks, a single frozen skill registry achieves the highest verdict-weighted correctness and reference consistency among the evaluated systems. Evolved skills raise correctness from 3.70 to 4.55 on our enhanced benchmark. In a separate 12-round operational study, only six of 33 proposed skills are promoted, while the monitoring non-correct rate falls from 20.0% to 12.5%. These results establish controlled skill scope, admission, and lifecycle management as the foundation for reliable self-improvement.

Adversarial patches to Vision-Language-Action (VLA) policies can cause both immediate action corruption and persistent state effects that remain after the patch is removed. Existing evaluations largely focus on continuous attacks and do not separate these two effects. We introduce a state-restoration protocol that removes the patch at matched action-chunk boundaries and measures subsequent recoverability under the same remaining step budget. Clean, random-patch, deviation-matched, and fixed-direction controls distinguish adversarial effects from occlusion, action-error magnitude, and directional persistence. We also evaluate a recovery adapter trained on attack-induced states under controlled intervention latency. On OpenVLA-OFT with EDPA attacks, only 36.2% of LIBERO-Long episodes remain recoverable after five chunks, compared with 89.9% and 87.0% for the deviation-matched and fixed-direction controls. Similar persistent effects are observed on autoregressive OpenVLA. The recovery adapter improves recovery from 7.7% to 47.4% at one-chunk latency, but its benefit decreases substantially with delayed intervention. These results show that adversarial effects can persist after patch removal and that timely intervention is critical for recovery.

Imitation-learned vision--language--action (VLA) foundation models acquire broad manipulation capabilities by scaling robot data across tasks and embodiments, but reliable deployment on a specific downstream task and hardware platform still requires post-training. Dexterous hands make this adaptation particularly difficult: their broad behavioural repertoire and high degree of freedom create a large and structured action space. Three obstacles are central: open-source VLAs do not natively provide an action interface for high-DoF hands; gesture mismatch during human-gated DAgger takeover creates command discontinuities and contaminates corrective trajectories; and reinforcement learning in the raw joint space is sample-inefficient. We present a unified four-step post-training pipeline comprising a learned temporal hand-action codec, supervised fine-tuning, DAgger, and real-world residual reinforcement learning. The codec adapts a pretrained VLA to absolute dexterous-hand commands. Buffered rollback, pose alignment, and smooth command blending enable continuous, task-relevant DAgger corrections, while latent residual RL confines exploration to coordinated hand motions captured by the codec. We evaluate the pipeline on five diverse real-world tasks spanning bimanual transfer, in-hand reorientation, and tool use. Within the reported post-training budgets, the resulting policies achieve 100\% success on every evaluated task over 20 trials per task. These results provide a practical path for adapting VLA foundation models to reliable real-world dexterous manipulation.

Runtime safety filters for learned manipulation policies typically define unsafe states as unions of object-wise keep-out regions. This representation can be unnecessarily restrictive for hazards that depend on a joint spatial relation, such as battery recycling, where a conductive payload can short a charged cell only when it approaches both terminals simultaneously. We study runtime filtering for this two-terminal hazard in LIBERO using frozen OpenVLA policies. We factor a runtime filter into three design choices: the predicate structure, its geometric margin, and the fallback action applied when a commanded action is rejected. We compare a conjunctive predicate, a conventional two-site keep-out, and a composite of the two. For each predicate, we vary its margin to obtain a frontier between task success and residual hazard. We then compare four fallback strategies at matched operating points: holding, retreat, sampled search, and a continuous-action barrier projection. Across three workcells, the three predicate families trace nearly identical safety--utility frontiers once each is evaluated over its own margin. In contrast, the fallback strategy has a substantially larger effect: holding reduces task success by up to 0.302 relative to retreat without reducing hazard, while both minimally invasive fallbacks leave substantially more residual hazard. This ordering transfers to a second policy and task suite, while retreat-based filtering remains effective under standing errors in the clearances available to the filter, although correlated error in the estimated payload size is more damaging than larger independent errors in terminal position. These results show that, for proximity-defined manipulation hazards, margin selection and fallback strategy can matter more than predicate structure in determining the safety--utility trade-off of a runtime filter.

Video agents have made substantial progress in long-video understanding. Yet effective video-agent systems require costly, time-consuming manual design and trial and error. Current self-improvement methods either refine low-impact prompts, recombine predefined micro-tools, or struggle with convergence in harness optimization. To bridge this gap, we target high-impact video-tool with VideoResearcher, a training-free multi-agent framework that autonomously designs, tests, and refines tools for video understanding, like a human researcher. VideoResearcher operates through dual Solving and Evolving loops: it analyzes tool-use trajectories to identify capability gaps, coordinates specialized agents to develop and validate executable tools, and reuses evolved tools to strengthen evidence acquisition in subsequent video reasoning. Through iterative tool refinement and validation, it progressively strengthens evidence acquisition without updating model parameters. VideoResearcher achieves state-of-the-art performance among self-improving agents and approaches the human-designed upper bound, demonstrating a training-free paradigm for long-video understanding that expands agent capabilities through autonomous tool development while reducing costly manual engineering.

Cross-view object geo-localization (CVOGL) typically assumes a fixed query image, overlooking the ability of mobile agents to actively acquire more informative observations. To address this limitation, we introduce Active Cross-View Object Geo-Localization (ActiveGeo), where an agent sequentially selects new viewpoints and determines when to stop, aiming to improve localization with minimal observations. We further propose ActiveMoPT, an ActiveGeo framework with three-stage training. First, Multi-View Prompt-Preserving Adaptation enables the model to aggregate multiple query views while reusing the initial prompt. Second, Trajectory-Guided Policy Initialization uses supervised agent trajectories to learn viewpoint selection and initial stopping behavior. Third, Cost-Aware Policy Refinement employs GRPO with a gain-cost reward to jointly optimize localization accuracy and observation efficiency. We also construct ActiveGeo-858, a zero-shot test set containing 858 scenes and 1,716 target annotations. Experiments show that ActiveMoPT achieves state-of-the-art performance on MoP-UAV using only 1.45 query views on average, and substantially outperforms previous CVOGL approaches under zero-shot evaluation on ActiveGeo-858.

Imitation-learned robot policies are frequently overfit to the visual conditions present in their training demonstrations. Consequently, variations in object color or background appearance often induce substantial performance degradation. A common mitigation strategy is to acquire additional demonstrations in each novel visual context; however, this approach is resource-intensive, requiring repeated access to a robot, a controlled environment, and human operation for every appearance condition to be covered. We introduce ReShoot, a framework that synthesizes visual diversity by re-rendering previously recorded demonstrations under altered appearances, thereby shifting the burden from data collection to generation. A vision-language model captions the scene, edits a targeted attribute (e.g., background, object color, or material), and an edge-conditioned video generator re-renders both camera views to match. The instruction is updated accordingly. The action sequence and proprioceptive trajectory are copied verbatim without relabeling, so each generated episode retains the recorded action and proprioceptive labels. On LIBERO, a policy trained on an equal mixture of recorded and re-rendered demonstrations matches the performance of recorded-only training (96.5% vs. 96.9%). Moreover, the mixed training set improves robustness to scene perturbations on LIBERO-Plus (85.5% vs. 82.3%). Across two physical robotic platforms, deploying ReShoot with 43 and 100 pre-collected demonstrations increased the success rate on recolored objects from 0.0% to 42.9% and 47.5%, respectively, while maintaining performance under the original recorded appearance.

Training embodied foundation models typically requires massive-scale datasets and extensive computational resources, yet often suffers from three critical limitations: (1) inefficient sample utilization due to low-informative samples; (2) imbalanced gradient contributions across heterogeneous tasks; and (3) severe credit assignment problem in long-horizon planning, where trajectory-level rewards indiscriminately penalize all tokens. To address these issues, we propose an efficient training paradigm that achieves state-of-the-art average performance through strategic data selection and hierarchical policy optimization. Our approach consists of three synergistic stages. First, Rejection Sampling-based Fine-Tuning (RSFT) filters out low-informative samples to establish robust behavioral priors while preventing distributional collapse. Second, Iterative Rejection GRPO (IR-GRPO) employs task-specific queues stratified by difficulty to keep datasets balanced across reinforcement learning iterations, coupled with a hybrid reward mechanism for precise cross-task feedback. Third, to enhance long-horizon task planning, we introduce Trie-GRPO, a novel reinforcement learning algorithm based on action prefix trees, which enables step-level advantage estimation. This resolves the credit assignment problem by isolating intermediate correct decisions from downstream errors, while effectively balancing exploration efficiency and depth compared to conventional search trees. As a result, EmbodiedMind achieves a state-of-the-art average performance of 70.02% across 18 benchmarks, and significantly outperforms other embodied foundation models in long-horizon task planning accuracy. Our project will be released for reproducibility.

Travel-planning agents generate itineraries that may become infeasible after acceptance because of flight cancellations, hotel unavailability, or attraction closures. Revising these itineraries involves full replanning, classical plan repair, and LLM-based travel-agent revision, whose differing task formulations and evaluation protocols hinder comparison. We conduct a systematic empirical study using two TREK-derived benchmark sets: 500 single-disruption cases, including feasible and infeasible instances, and 200 feasible simultaneous compound-disruption cases. We compare LLM-Z3 full replanning, IPyHOPPER hierarchical repair, and an iTIMO local-revision adapter across effectiveness, plan stability, and computational cost. LLM-Z3 with Gemini achieved the highest observed compound-disruption success. IPyHOPPER nearly matched that configuration's single-disruption overall success, while preserving substantially more of the accepted itinerary on successful repairs. Successful hierarchical and local repairs made fewer edits and retained more accepted commitments than full replanning. Computational profiles differed: IPyHOPPER used no LLM inference, the evaluated LLM-Z3 adapter used compact one-call inference, and the iTIMO adapter consumed substantially more tokens. The study provides practical guidelines for balancing feasibility recovery, commitment preservation, and computational cost within evaluated settings.

This paper proposes a Retrieval-Augmented Generation (RAG) framework for scientific image quality assessment, designed to simultaneously address both the understanding track (SIQA-U) and the scoring track (SIQA-S) of the SIQA challenge. We construct a multimodal index that integrates textual semantics with fine-grained visual features, and develop a multi-route retrieval and fusion mechanism to provide large language models with highly relevant reference cases, thereby enhancing their capability to evaluate complex scientific images. Experimental results demonstrate that the proposed framework effectively aligns with the judgment criteria of human experts. Ultimately, our method achieves 1st place in the SIQA-U track of the SIQA challenge at the ICME 2026 Grand Challenges.

Large language models (LLMs) are increasingly integrated into vehicle voice assistants. But linking natural-language requests to vehicle functions creates a safety-critical authorization problem. Before executing a command, the system must choose whether to execute, refuse, clarify, require confirmation, defer to manual control, trigger an emergency response, or make no tool call. To our knowledge, prior evaluations do not isolate this pre-action decision across speaker role, authentication status, vehicle state, and tool availability. We introduce a 202-scenario benchmark with Reference Decisions under a seven-class taxonomy. We evaluate two local open-weight models and three API-based LLMs using Decision Alignment and safety-specific error metrics. Alignment ranges from 40.1% for Llama 3.2 3B to 89.1% for Gemini 3.1 Pro Preview. The API-based models score between 83.2% and 89.1%, with no statistically significant differences among them. Even these models produce two to three False Executes among 161 non-execution scenarios, and persistent errors remain in confirmation and manual-control decisions. A controlled Llama 3.2 3B ablation increases alignment to 40.1% under the structured authorization policy, versus 28.2-29.2% under schema-only and generic-safety baselines, but it does not eliminate False Executes. Structured LLM decisions are therefore insufficient as a standalone safety mechanism, and deployment requires an independent enforcement layer that verifies tool permissions and vehicle-state constraints before invoking any vehicle function.

We present VGGT-GS SLAM, a monocular 3D Gaussian Splatting SLAM system designed for uncalibrated videos. Starting from feed-forward VGGT pose and depth priors, our system performs submap differentiable bundle adjustment that jointly refines camera poses and a 3D Gaussian map, while optimizing submap-shared intrinsics and radial--tangential distortion through analytic calibration Jacobians. To improve global consistency, we introduce Gaussian-native alignment (GNA) for camera-anchored scale refinement between sequential submaps and verification of loop-closure candidates. Extensive experiments on standard indoor benchmarks show consistent improvements in localization accuracy and strong rendering quality under uncalibrated settings, establishing a strong baseline for uncalibrated Gaussian SLAM.

Modern applications generate massive volumes of raw telemetry data, but translating those noisy, heterogeneous event streams into actionable business insights remains a fundamental challenge. Data engineers and analysts expend substantial effort reconciling semantic discrepancies, hand-crafting parsing logics, and maintaining fragile mappings between raw data and business KPIs. In this paper, we present an end-to-end framework that fully automates the construction of a business semantic layer from application raw logs. Our approach introduces a two-stage semantic abstraction: first, high-level business features are identified via LLM inference augmented with domain-specific industry knowledge; second, fine-grained business nodes are derived through a structured pipeline comprising data refinement, hybrid retrieval, multi-stage filtering, semantic clustering, and canonical naming. Evaluation on production-scale telemetry demonstrates that our system improves human-assessed semantic quality from 50 to 80+ on a 100-point scale, reduces maintenance effort by 80%, filters out 74% of noise, and achieves 0.87 Cohen's kappa via an integrated LLM-as-Judge evaluation, enabling continuous, scalable quality assurance. Overall, our work distinguishes itself from prior work by addressing the novel problem of business semantic layer induction from raw telemetry, operating without labeled training data or manual rule engineering.

Dexterous food manipulation requires control under deformation, occlusion, and uncertain contact. We present TacSushi, a tactile-grounded, Cosmos3-based world-action policy that learns from recorded future consequences while acting on current observations. The backbone encodes current RGB, language, and hand state, and feature-wise gated fusion incorporates fingertip tactile features into the action representation. During training, a decoder conditioned on demonstrated action chunks predicts logged future visual observations, task progress, relative contact risk, and tactile summaries; this decoder is removed at deployment. Failed trials provide consequence supervision, but their actions are excluded from imitation. We train TacSushi on 340 successful and 50 failed real-robot trials and compare six methods in 600 separate rollouts across three in-distribution tasks and two out-of-distribution ingredient variants. To assess food quality beyond a single geometric threshold, we score terminal outcomes using an anchored visual-quality protocol that equally weights five human ratings and three vision-language-model ratings per rollout. Full TacSushi achieves 68.3% average in-distribution success and 37.5% out-of-distribution success, compared with 36.7%/10.0% without future-consequence supervision and 25.0%/17.5% with direct tactile concatenation in place of gated fusion. These comparisons support complementary benefits of feature-wise gated tactile fusion and training-only predictive supervision.

This empirical study is an independent reproduction of the dissociation Zhao reported in 2026. The shape of a large language model's chain-of-thought entropy trajectory predicts whether the final answer is correct, while the magnitude of its total entropy drop does not. The dissociation merits reproduction because the magnitude half rests on a single 300-problem run with one model at one seed, while the shape half was reported at full scale on both benchmarks and on a second model family. Registered at OSF before any confirmatory run, the reproduction crosses the complete GSM8K and MATH-500 benchmark test sets with four open-weight models including one reasoning-distilled model of a kind the original did not test. The shape signal replicates. The magnitude signal divides by setting. On the anchor model the accuracy gap between monotone and non-monotone chains is +9.6 percentage points on GSM8K and +27.5 on MATH-500, while the rank correlation of the total entropy drop with correctness is -0.018 on GSM8K and +0.414 on MATH-500. On the reasoning-distilled model the binary form of the shape signal fires on about one chain in a hundred, too few to estimate the registered contrast, while the graded violation count remains predictive there. In an exploratory comparison the final-step entropy alone outperforms the binary shape flag in all eight model-by-benchmark cells by ROC area, and in six or seven by the risk-coverage area the original reports, depending on an integration range the original does not state. The study contributes a reproduction of the shape signal at full test-set scale under seven documented protocol differences, a map of the settings where the magnitude signal holds and fails, and measurements of four protocol dependencies the original does not report.

Adapting robots to new objects and tasks requires interaction experience that can be costly to obtain. We present WorldContact, a contact-centric world model for deformable-object manipulation, constructed from a limited set of high-quality trajectories to generate additional training data efficiently. It predicts object dynamics using larger time steps than the source numerical simulator, which requires small integration steps to resolve rapid motion and prevent interpenetration. We evaluate WorldContact across 16 shopping-bag manipulation tasks. State-rollout measurements on a single H100 GPU show a $10\times$ speedup over the source simulator, excluding rendering and disk I/O. We use the generated data to fine-tune an existing vision-language-action policy and deploy it directly on a real robot. In bag lifting, the same policy achieves 65% single-attempt success when fine-tuned on source simulation data alone, compared with 95% when fine-tuned on the dataset expanded with WorldContact. These results support efficient data generation with WorldContact for robot policy adaptation.

This study developed and evaluated a deep-learning-based perception framework for selective robotic cotton picking. The dataset contained 1,008 annotated field images collected using three cameras under varying natural lighting and weather conditions. Object-detection models from the YOLOv8 through YOLOv13 families were evaluated using their default configurations, while segmentation performance was assessed using YOLOv8-seg, YOLOv11-seg, YOLOv12-seg, the Segment Anything Model (SAM), SAMv2.1, FastSAM, and Grounded-SAM with the Recognize Anything Model (RAM). Among the detection models, GELAN-s achieved the most favorable balance between mean average precision (mAP) and inference speed, obtaining an mAP of 86.1%, precision of 81.6%, recall of 76.6%, and an F1-score of 79.0%, with an average inference time of 42.3 ms per image. Among the direct segmentation models, YOLOv12-m-seg provided the most favorable balance between [email protected] and FPS, achieving a segmentation [email protected] of 83.7% with an inference time of 20.4 ms per image. In the detection-prompted segmentation approach, bounding-box prompts generated by GELAN-s improved the localization of cotton bolls for SAM and SAMv2.1, while SAMv2.1 Tiny consistently outperformed FastSAM and Grounded-SAM with RAM. In the area-based evaluation against manually annotated segmentation masks, YOLOv12-m-seg achieved an $R^2$ value of 0.966, compared with 0.860 for GELAN-s + SAMv2.1 Tiny. Field experiments conducted using a UR5e robotic manipulator, a custom end-effector, and a ZED2i stereo camera further validated the effectiveness of the YOLOv12-m-seg model for real-time cotton boll detection, segmentation, and selective picking under varying confidence levels. These results demonstrate that YOLOv12-m-seg provides an efficient perception model for robotic cotton harvesting and has strong potential for field deployment.

Data attribution methods using gradient similarity are widely used to analyze and select training data for large language models, but what gradient similarity actually measures is debated. Some interpret it as identifying task-relevant skills, while other work reports that surface form is the main factor. We resolve this debate for supervised fine-tuning examples by varying task and answer format independently. Specifically, we render benchmarks in different answer formats, such that datasets can share a task without a format or a format without a task. We find that gradient alignment follows the answer format, as benchmark pairs sharing an answer format align strongly (disattenuated cosine near 0.4), while same benchmarks rendered with different answer format classes show no alignment (near 0.0). We demonstrate that this ordering holds from the earliest pretraining checkpoints through post-training, and across model scales and families. We then analyze the released selections of LESS, a gradient-based data selection method for instruction tuning, and find that each target's selections over-represent the target's own answer format. Hence, we demonstrate that gradient-based attribution methods track format similarity more than task semantics, meaning that such methods, as well as the semantic interpretation of the gradient, should be tested on data where answer format and task vary independently for greater robustness and reliability.

Mechanical intelligence, loosely defined as the reduction in control burden afforded by a robot's physical form, has become a prominent concept in robotics, with instantiations in bioinspired robotics, soft robotics, robotic swarms, and many other areas. However, rigorous theoretical understanding and quantitative measures of mechanical intelligence have lagged behind the engineering systems that the community has developed. In this work, using modern legged robots as a benchmark and exemplar, we propose several information-theoretic metrics for quantifying mechanical intelligence. By viewing body dynamics as both a computational process and a communication channel, we show that several prior insights in legged-robot engineering can be described using information theory, and we quantify how bits are processed by mechanical modes and across robot coordinates. Specifically, we examine the trade-off between explicitly incorporating compliance through series-elastic actuation and using so-called proprioceptive, low-gear-ratio transmissions, and we explore how these mechanisms interact with control policies during locomotion. We develop these results on systems of increasing complexity: a simplified linear model of a robot-leg transmission, a nonlinear single-leg simulation, and simulated quadruped robots controlled by a learned policy while navigating challenging terrain. These results lay the groundwork for broader study of robot mechanisms and their role in embodied computation.

To keep coding agents from going off the rails, production systems now review each proposed action with a blocking monitor that can reject it before it runs (Auto Mode in Claude Code, Guardian in OpenAI's Codex). Prior evaluations of such monitors largely measure robustness to accidental harm or prompt injections from untrusted sources looking to hijack the agent. Less understood is how they hold up when the agent they monitor is persistently misaligned. To understand this risk, we task an adversarial agent with evading production blocking monitors and causing catastrophic harm, e.g. by exfiltrating its own weights. We find that when instructed with high-level attack strategies, adversarial agents can succeed through several distinct mechanisms, such as agent-generated prompt injection against the monitor, multi-agent attacks, and malicious compaction. In particular we find that in 79% of trials, the agent can use an injection attack against Auto Mode and Guardian to run arbitrary bash commands. We also find that it is possible to greatly improve Auto Mode through design changes like enhancements to tool coverage, transcript formatting and an agentic monitor stage. Even with these improvements, preventing multi-context attacks at an acceptable cost remains an open problem. By detailing our red-teaming methodology and highlighting new attack vectors, we aim to help defenders evaluate their mitigations against the possibility of persistent malign coding agents. Code is available at https://github.com/safety-research/red-teaming-auto-mode.

In mental health care, reasoning over patient journeys is a key task for clinicians. Yet these journeys, encompassing a longitudinal progression of biological, psychological, and social events, are often spread across disparate unstructured text narratives, making temporal recovery challenging. We present CliniCIRCA, a multi-stage LLM framework for Calendar-anchored, Imprecision-aware Reconstruction of Clinical Annals. To our knowledge, CliniCIRCA is the first to temporally classify clinical events across unstructured discharge summaries without event-level timestamps. From 14,882 MIMIC-III mental health admissions, we first construct a benchmark of 52 discharge summaries on which CliniCIRCA produces 15,891 temporally tagged events. After correcting 629 errors based on a clinician-in-the-loop evaluation, we produce verified gold-standard labels. Finally, the corrected timelines drive a temporally grounded summarization stage that compresses each source 1.52 times into a date-grouped chronological record. We then scale the framework to generate 1,000 silver-standard timelines and evaluate them as training data. Compared with zero- and few-shot prompting, instruction tuning generally improves five open-weight models on event extraction, temporal tagging, and summarization across silver and clinician-verified evaluations.

Assembly, wear, and component replacement perturb the sensor extrinsics and joint zeros encoded by a humanoid CAD model. Existing procedures calibrate one sensor pair or require external fiducials. Using only robot-native motion and onboard sensing, we present OmniCalib, a target-free workflow that calibrates the full upper limbs---all 14 arm joint zeros and the extrinsics of both wrist and chest cameras---as well as lower limbs and the multi-camera head rig. Each module matches a robot-native task to a parameter block, checks observability, and writes only supported corrections to the CAD model. Our depth ICP method recovers all 14 arm joint zeros and calibrates all RGB-D camera extrinsics without any calibration target. Relative to CAD, the estimated extrinsic corrections are 10.56 mm and 1.74 degrees for the left wrist, 6.33 mm and 1.25 degrees for the right wrist, and 9.81 mm and 0.929 degrees for the chest RGB-D camera. ICP point-to-plane residual is 2.09 mm. On the same injected offsets, ICP and ArUco recover all 14 joint zeros below the 0.1-degree encoder-resolution reference. On an AGIBOT A3 Ultra humanoid, four static double-support stances recover all 12 lower-limb joint-zero offsets injected with an RMS error of 0.063 degrees. The head module combines multi-camera visual odometry with legged odometry and dynamic compensation through the live ROS transform tree. Using only planar walking, it attains a mean SO(3) error of 1.061 degrees across three sequences. The best sequence reaches 0.775 degrees, competitive with iKalibr at 0.902 degrees from rich 6-DOF excitation. Rig-relative angles repeat within 0.140 degrees. Injection recovery and held-out tests validate each observable block.

Vision-language-action (VLA) models let robots follow language instructions, but their language backbones of several billion parameters are the main obstacle to running them on robot hardware. Structured pruning reduces that backbone, and removing 63% of it from OpenVLA-OFT drops LIBERO-Long success from 93.2% to 0.8%. A recent approach restores such a model with supervised fine-tuning followed by reinforcement learning, which needs online rollouts and hundreds of GPU-hours. We recover most of the lost success entirely offline. Width pruning narrows the blocks but keeps the residual stream at its original size, so teacher and student hidden states have the same shape and are matched directly, without a projector. Training against a cache built in one teacher pass lifts the 63%-reduced student to within 3.5 points of the teacher in about 8 GPU-hours. A sweep over nine ratios locates where the recovery objective starts to matter. Up to 45% reduction the two do not differ significantly on OpenVLA-OFT. Hidden-state distillation then adds +2.1 to +4.5 points there between 63% and 87%, and +9.4 to +22.1 points on CogACT from 63% onward. At 81% on CogACT, a tripled recovery budget narrows the distilled student's gap to the teacher to 3.9 points on average, while supervised recovery stays more than 20 points below. At matched compression, width pruning yields higher success and depth pruning lower latency. On a 6-DoF manipulator, the distilled student at 72% reduction reaches 77.5% success against 59.5% for supervised recovery, runs 2.23x faster on-board than the teacher, and uses 62% less memory.

Teleoperation, a human-in-the-loop control scheme, allows a human operator to remotely command and guide a mobile robot to navigate in off-road environments, yet fluent and user-friendly tele-navigation requires an alignment of traversability evaluation between human and robot. In the teleoperation system, the human operator typically utilizes off-site incomplete and delayed feedback via a human-machine interface to make a judgment of traversability, while the robot makes such an evaluation based on in situ onboard sensory information, which could cause divergent traversability estimation and thus generate mismatched decisions and actions. Existing approaches for traversability modeling, estimation, and prediction are mainly derived from the view of robots, i.e., robot-centric, and are practically suitable for fully autonomous mobile robots, but neglect the influence of human operators. To address this problem, this paper extends the concept of traversability from robot-centric to human-centric by accounting for the operator's cognitive states, such as attention, workload, and risk tolerance or awareness, termed tele-traversability. We first revisit the definitions and roles of traversability in robotics and then extend them to teleoperation settings. Finally, we highlight future trends and open challenges of tele-traversability toward human-centric teleoperation systems.

Disease severity classification for genetic conditions is subjective and labor-intensive, creating bottlenecks in genomic screening, where commercial panels vary widely in size and overlap. We developed an autonomous AI agent integrating Reasoning and Acting (ReAct) with Retrieval-Augmented Generation (RAG) to classify 10,211 Human Phenotype Ontology terms. It uses American College of Medical Genetics (ACMG)-endorsed severity guidelines and American College of Obstetricians and Gynecologists (ACOG) quality-of-life criteria to retrieve PubMed literature, generate interpretable reasoning chains, and independently verify claims. At the phenotype level, using expert-curated cohorts, the agent achieved 93.55% accuracy (MCC 0.9237) with 82.6% to 91.4% of claims supported by direct evidence or valid inferences. Gene-level severity was aggregated across 8,738 pairs, identifying 3,283 autosomal recessive pairs with severe or profound presentations. External validation showed 95.2% concordance with Mackenzie's Mission gene list. This system enables standardized panel design by providing reliable, automated classification supported by direct evidence.

Spatial intelligence requires more than describing object locations. Under incomplete observation, models must identify and acquire missing evidence, interpret it in a common spatial frame, and act on it. We introduce VA-Bench to evaluate the complete observe-reason-act-revise loop. General-purpose MLLMs learn procedural context from RGB-only demonstrations, actively select camera viewpoints, issue metric Cartesian commands, and revise them from execution feedback. Models receive no privileged object poses, oracle trajectories, or learned action heads. A fixed model-agnostic controller executes only model-specified targets. VA-Bench contains 14 base task families (11 single-arm and three dual-arm), seven held-out geometry/layout variants, and a long-horizon five-object composition track. We evaluate 12 primary model conditions in three independent runs over the same 20 physically verified seeds per base task, reporting terminal success, nine trajectory-level behavioral diagnostics, and subtask progress. First, the best-performing model scores 100.0% on target localization and 78.9% on spatial relations in the annotated run. Its three-run macro-average task success is only 53.93+/-3.17%. Second, active camera control significantly improves task success over passive multi-view observation. In one matched comparison, success rises from 27.86% to 57.50%. Third, held-out geometric transfer can reduce task success by over 30 percentage points. No model completes a strict long-horizon episode, despite substantial partial progress. VA-Bench thus tests whether general-purpose MLLMs can turn visual demonstrations and actively acquired evidence into successful embodied action.

Model fusion integrates the capabilities from source models into a single target model. As of June 2026, Hugging Face hosts more than 2M models. This growing pool provides a rich base for model reuse and capability integration. Yet existing surveys often cover only separate parts of this space, and they do not provide a unified definition or a systematic taxonomy. This survey defines model fusion and organizes prior work into three levels: parameter-level, representation-level, and behavior-level fusion. We also review related metrics, benchmarks, and applications, summarize current challenges, and identify future directions. Our goal is to provide a clear map of this area and support future work on model fusion. A comprehensive list of papers about model fusion is available at https://github.com/Baicaihaochi/Awesome-Model-Fusion-Survey.

Open-vocabulary segmentation enables rich semantic perception for UAVs, but frame-wise predictions can remain temporally inconsistent across repeated observations and changing viewpoints. We present PerSeM, a training-free persistent semantic memory framework for long-horizon open-vocabulary UAV mapping. PerSeM associates frame-wise semantic observations with persistent world-space voxels and constructs a majority-based semantic memory, which is conservatively refined through history-preserving spatial refinement, trust-aware replay, and context-guided verification. Experiments on the Forest and UAVScenes benchmarks show that persistent 3D memory provides substantial gains in semantic correctness and temporal stability over frame-wise predictions. Beyond this strong persistent-memory baseline, PerSeM provides consistent additional improvements, improving both semantic accuracy and temporal stability across all five evaluated UAVScenes sequences. Analysis using regions identified independently of the final PerSeM predictions further shows that these gains are concentrated in semantically difficult and temporally unstable regions, where majority-based memory is most likely to remain uncertain. These results demonstrate that persistent 3D aggregation provides a strong foundation for long-horizon semantic mapping, while conservative refinement of uncertain memory states can provide additional improvements without retraining or additional neural-network inference.

Low-altitude wireless networks (LAWNs) are emerging as a key infrastructure for heterogeneous unmanned aerial systems that support concurrent services within a shared three-dimensional airspace. Their coexistence creates strong coupling among mobility, connectivity, and shared network resources, while heterogeneous services impose distinct and time-varying requirements. These interactions naturally form a dynamic non-cooperative game in which both operating conditions and coordination objectives evolve over time. Conventional optimization and learning-based controllers typically rely on predefined objectives, limiting their ability to adapt autonomously to changing service requirements and resource priorities. To address this challenge, we propose a hierarchical hybrid large language model (LLM)- multi-agent reinforcement learning (MARL) architecture organized as a dual-loop structure. Specifically, an outer adaptation loop employs LLM-assisted game orchestration to interpret service requirements and operator intent, and reconfigure objectives and resource priorities, while an inner loop executes decentralized, parameter-conditioned MARL policies under the configured game. A logistics-monitoring case study illustrates how the proposed framework facilitates coordinated coexistence among heterogeneous services, adapting to evolving operating conditions without retraining the underlying MARL policies. Finally, we discuss key challenges and research directions toward scalable, trustworthy, and adaptive agentic LAWNs.

Simultaneous localization and mapping (SLAM) is one of the services running on an autonomous robot. It is typically run to assist other tasks such as planning, manipulation, etc. All these tasks are run on edge hardware and are subject to severe resource constraints. However, most SLAM systems are built and tested in isolation, and their performance is reported as if they are the only task running on a system. We observe that existing benchmarks lack a common mechanism for comparing SLAM systems under realistic resource constraints. To address this limitation, we have developed SLAMSqueezeBench, a framework that allows testing of SLAM systems under realistic workloads on edge hardware. It does so by imposing constraints on compute and memory resources available for the SLAM system during execution. It also simulates realistic camera frame acquisition with frame drops when a finite buffer is full. Using SLAMSqueezeBench, we compare nine SLAM systems spanning classical systems, learning-based systems, and approaches for Gaussian splatting. Our testing framework will be available for use by the community upon publication.

Air-ground transportation research increasingly relies on co-simulation, yet constructing scenarios remains labor-intensive and difficult to validate. More importantly, a generated scenario may execute successfully while failing to realize the spatial, temporal, communication, or behavioral relationships requested by the user. This paper presents AURORA, a natural-language-driven agentic framework that treats air-ground scenario generation as a process of compilation with verification. Central to AURORA is the Air-Ground Scenario Graph (AGSG), a typed intermediate representation that explicitly connects agents, aerial missions, events, communication links, success conditions, and their cross-domain dependencies. This shared representation enables simulator-grounded parsing, joint road-airspace grounding, temporal planning, pre-execution feasibility checking, trace-based runtime verification, failure localization, and bounded repair within a unified workflow. We further introduce AURORA-Bench to evaluate not only whether generated scenarios execute, but whether they faithfully realize the requested interactions. Experiments across multiple language models show that structured execution substantially improves reliability, while runtime verification exposes silent failures that completion-based evaluation overlooks. Localized repair further resolves many violations without regenerating the entire scenario. The results show that reliable scenario generation requires verifying realized behavior, not merely executable code, and demonstrate the value of explicit intermediate representations for verifiable and repairable language-driven co-simulation.

Web agents often revisit the same sites, yet most evaluations discard the procedures learned in earlier successful interactions. We introduce EconSkills, a skill library and evaluation framework that distills verified EconWebArena trajectories into parameterized standard operating procedures for retrieving live economic data. Each skill records its scope, navigation procedure, site-specific guidance, verification checks, and recovery steps while replacing source-instance values with placeholders. EconSkills separates two questions: whether a known relevant procedure transfers to a held-out task, and whether an agent can retain that benefit when selecting from a library. In controlled transfer, matched skills improve success over no-skill prompting and require fewer steps on paired successes, while abstraction is substantially more effective than replaying raw trajectories. At library scale, retrieval is competitive with the no-skill baseline overall and performs best on directly covered tasks; coverage-stratified outcomes show that approximate matches on uncovered tasks offset these gains. Browser trajectories further identify when procedural guidance shortens portal-specific navigation and when semantic verification remains necessary. These results establish that reusable economic web procedures can transfer across task instances and provide a concrete design target for coverage-aware selection and context delivery.

A fundamental assumption in robotic perception is that the sensor's field of view (FoV) is fixed relative to the robot body. Motion-decoupled sensors, such as gimbal-mounted cameras and MEMS-based LiDARs, instead allow sensing direction to be controlled independently at runtime. This freedom creates a computational challenge: efficiently selecting useful viewing directions online in feature-dense environments. We propose PIVOT, a lightweight iterative method that optimizes sensor viewing direction along a fixed translation trajectory to maximize feature visibility. Under a conical FoV model, visibility depends only on the optical axis, yielding a two-degree-of-freedom optimization on the viewing sphere $S^2$. Coordinate-free $SO(3)$ exponential-map updates enable efficient continuous optimization without explicit angular parameterizations or exhaustive viewing-sphere search. Monte Carlo evaluations retain 98.1--99.6% of brute-force visibility with a 76--85x speedup. Photorealistic simulation and real-world experiments further demonstrate improved visual localization robustness and practical viewpoint control on a quadruped robot.

As model-generated content is increasingly consumed by other model instances in automated workflows, a practically important question arises: can a model embed a signal in natural language that an independent instance of the same model can detect, relying only on shared pre-training and task instructions, without any shared memory or coordination-specific training? We introduce For Your Eyes Only, a cooperative signalling game designed to evaluate this directly. A Sender produces free-form descriptions for two words, one of which is a hidden target; an isolated Receiver must identify it. We evaluate seven contemporary models from four architectural families on 300 word pairs from established psycholinguistic corpora, using the Double-Pass Success Rate to control for output biases. We find that most models struggle to maintain coordination once they are required to avoid detectable signals, while one frontier model retains near-perfect performance even after such filtering. We further show that models can direct this capability toward deliberate misdirection, and that coordination is consistently weaker across architectures than within them.

Agents can now externalize experience into memory, consolidating historical traces into semantic knowledge and procedural shortcuts that persist between sessions. Such memory is typically private to a single agent. We argue that agentic memory benefits from being collective, because trustworthy knowledge of the shared environment---the data sources, services, and tools agents depend on---cannot be established by any single agent, only corroborated across many independent observers. We present Cairn, a community reputation platform that captures collective knowledge, allowing agents to query the community's opinion of a resource before use and to submit evidence-backed ratings afterward. Cairn aggregates observations via a time-decayed Beta model with confidence shrinkage and supports semantic discovery over reviewer rationales. We evaluate Cairn's reputation engine under adversarial simulation (e.g., lying, collusion, camouflage), benchmark its retrieval performance, and report a case study of rating heterogeneous agents in production.

Text-based person retrieval under a sim-to-real gap (synthetic training data, a real-image gallery) is usually tackled with costly fine-tuned cross-encoders. We ask whether a frozen-encoder system can compete. We present SCOUT, which casts cross-modal retrieval as prediction in embedding space. A trainable predictor maps the patch tokens of a frozen video encoder into the embedding space of a frozen text encoder under a bidirectional InfoNCE objective, and no encoder is fine-tuned in the base model. The video encoder is V-JEPA, the text encoder is EmbeddingGemma, and the predictor is initialized from a Qwen3.5-0.8B decoder. We make three findings. First, the best frozen text encoder is simply the one whose geometry best matches the video features. A training-free alignment score ranks three candidate text encoders in the same order as their retrieval accuracy on our held-out split (Spearman $ρ= 1.0$); a fourth, LLM-based encoder shows the rule is metric-dependent, holding for a neighborhood-overlap score ($ρ= 0.8$) but not for a linear probe ($ρ= -0.2$). Second, two precision-targeted levers, parameter-efficient ExPLoRA adaptation of the video encoder and a training-free attribute-decomposed reranker built on a vision-language model, improve the top-rank precision that otherwise limits the frozen system, adding 2.2 points of leaderboard R@1. Third, a local-versus-public calibration study explains which interventions transfer to the real domain. On AI City Challenge 2026 Track 4 the full retrieve-fuse-rerank system reaches 84.25 mAP@10 on the final leaderboard, while a single frozen model submitted alone reaches 60.63. Our trained components cost about 95 GPU-hours. CMP, the dataset authors' fine-tuned cross-encoder that trains for sixteen GPU-days, is one fusion member of the full system, not an alternative. Code and annotations: https://github.com/abtraore/SCOUT-ECCV

The muscles that power insect flight fall into one of two categories: 1) synchronous muscles that contract under direct control from the nervous system, and 2) asynchronous muscles which have an intrinsic stretch activation response that spontaneously generates wingbeats without the need for signaling from the brain. It is thought that the emergent nature of asynchronous wingbeats provides both adaptive and responsive capabilities for flight control. To date, most flying robots use synchronous actuation. In this paper we develop the first flight-capable flapping wing robot that uses asynchronous actuation. We demonstrate that asynchronous actuation allows wings to respond to changes in the resonant mechanics of the body without control input, and wings can react instantaneously to collisions with obstacles with no extrinsic sensing needed. Flight tests within cluttered environments demonstrate that asynchronous actuation significantly improves stability and performance when compared to synchronous actuation. In total this work demonstrates that a flapping wing robot actuation strategy that emulates the asynchronous muscles of flying insects can provide fast, reactive actuation responses before a control system would need to intervene. This partitioning of embodied control to both the low-level actuation dynamics and and high-level sensorimotor system provides a compelling blueprint for new flying robots.

We present ParticleSplat, a self-supervised object-centric learning method that decomposes scenes into a set of latent ''particles'' representing semantic entities through feedforward 3D Gaussian Splatting. Building on the Deep Latent Particles (DLP) framework, which represents images as a set of particles with attributes such as position, scale, and visual appearance, we address a key limitation of DLP: its inherently 2D nature, which prevents explicit 3D spatial and geometric reasoning that are critical for downstream tasks such as robotic manipulation. Leveraging the structural similarity between latent particles and 3D Gaussian primitives, we introduce a 3D latent particle space trained with a novel view synthesis objective. Our model jointly encodes multiple views with camera poses into a shared 3D object-centric latent space, then transforms particles into particle-aligned 3D Gaussians whose composition reconstructs the full scene. On simulated and real-world datasets, we show that this formulation inherently learns object masks without supervision and supports controllable 3D scene editing, such as moving objects by modifying particles in the latent space. We further establish that the learned 3D representation improves downstream performance on robotic manipulation tasks.

Robots must complete their tasks and maintain the achieved outcomes while avoiding safety failures at all times. Strict reach-avoid-stay (sRAS) formalizes this requirement: safely reaching a target and remaining there indefinitely after first entry. We propose an sRAS Q-control barrier function (CBF) safety filter for high-dimensional black-box systems under bounded uncertainty. Our construction combines a stay value encoding safe permanent residence in a target subset with a reach-avoid value encoding safe reachability of this subset while avoiding target states from which safe permanent residence cannot be guaranteed. We prove that these values jointly yield a valid robust discrete-time CBF and lift them to state-action Q-functions for runtime intervention. For exact values and under a measure-zero condition, our filter preserves sRAS feasibility from almost every winnable initial state and keeps the system safely within the target after first entry, against all admissible uncertainty realizations. We adopt reachability-based adversarial reinforcement learning for scalable value approximation using only black-box interactions. Notably, neither synthesis nor deployment of our filter requires known dynamics, affine structure, value derivatives, or hand-designed barriers. We validate our framework in quadruped gap jumping in simulation and hardware, where the robot crosses the gap, lands safely, and remains safe afterward. Simulated F1TENTH races further demonstrate safe overtaking and lead retention.

Graph-based retrieval-augmented generation (RAG) is widely used for multimodal, cross-document question answering. However, building corpus-level graphs is expensive, slow to query, and difficult to maintain. We present TrioRAG, a graph-free multimodal framework that integrates evidence from three complementary signals: the question, the anchor image, and a VLM-enhanced query generated from both. Each signal retrieves independently over a shared multi-vector index of page text and page images, and the results are combined through late fusion. Further, we introduce AutoQA, a multimodal automotive benchmark whose questions are grounded in noisy, web-sourced images rather than clean document-sourced figures. Its questions require reasoning across manuals. We position it as a model-curated testbed rather than a human-validated gold standard. Across three benchmarks, TrioRAG matches or outperforms graph-based systems while reducing total cost and accelerating per-query inference by 1.6-2.3 times. By construction, AutoQA grounds its questions in out-of-corpus web images. In this setting image retrieval reaches only 19.3% document-level recall, while text-derived signals, especially the VLM-enhanced query, keep retrieval robust.

We investigate static-to-dynamic electromagnetic field conversion in a temporally switched Lorentz medium initially subjected to an electrostatic field. An abrupt change of the oscillator strength drives the material away from its initial equilibrium, while the polarization retains memory of the pre-switch state and acts as the source of the resulting transient. We formulate the resulting initial-value problem in the Laplace domain and derive an analytical representation of the generated field in terms of the poles of the finite Lorentz slab. Material dispersion gives rise to a richer modal structure than in the nondispersive case, including families of slab-mode poles that accumulate toward the singularities associated with the Lorentz polarization dynamics. We further show that material memory governs the earliest stage of the transient, producing a smooth field onset and setting the velocity of the earliest propagating disturbance through the high-frequency permittivity. Because the temporal switch generates broadband spectral content, the transient samples positive-, near-zero-, and negative-permittivity regions of the post-switch material response, leading to qualitatively different spatial field distributions. These results reveal how material dispersion and memory fundamentally shape the generation and evolution of radiation produced from an initially static electromagnetic state.

The increasing complexity of modern power systems, driven by high renewable penetration, load variability, and operational uncertainty, demands fast and reliable solutions to the AC optimal power flow problem (AC-OPF). Traditional optimization methods, though accurate, often struggle with scalability and high computational burdens, making them impractical for real-time use in large networks. This paper introduces PowerModels-ACOPF-AI, a two-stage Bayesian Neural Network (BNN) surrogate designed to predict generator set points, bus voltages, and phase angles with uncertainty. The framework integrates a performance-sensitive on-the-fly learning mechanism that identifies regions of degraded prediction accuracy and dynamically retrains with additional AC-OPF solutions generated by PowerModels.jl. This self-adaptive loop ensures robust performance, enabling the model to maintain accuracy under novel or highly variable operating conditions. Validation on benchmark test systems of different sizes, namely the 30-bus, 200-bus, and 500-bus networks, demonstrates strong generalization capability, efficient handling of stochastic renewable injections, and the ability to provide rapid, uncertainty-aware predictions. Beyond predictive accuracy, the proposed approach offers practical value as both a real-time advisory tool for system operators and a fast initializer for conventional solvers, thus supporting resilient and efficient grid operation in future power systems.

Robotic additive manufacturing (AM) extends material-extrusion printing beyond gantry kinematics but makes process planning robot-dependent. A slicer-generated plan that appears favorable in part coordinates can become infeasible or robotically unfavorable on a manipulator because slicer-process decisions and part orientation determine the generated path, while part orientation and workspace placement affect its kinematic realization. Existing AM tools, large language model (LLM)-based decision-support methods, and digital-shadow systems do not provide integrated pre-execution evaluation of these coupled decisions. This paper presents agentic robotic additive manufacturing (A-RAM), an agent-specialist-tool framework that converts user intent and a part file into traceable, execution-ready plans. The LLM interprets manufacturing objectives and constraints, identifies prescribed and searchable planning variables, and encodes this reasoning in a schema-constrained request; a deterministic Planning Agent instantiates the corresponding search workflow, while domain tools compute quantitative evidence for slicing, placement, inverse kinematics, trajectory timing, Joint-6 jerk, and extrusion. The framework is evaluated on a six-axis robotic-arm AM cell through three case studies covering expert-specified planning, goal-only planning, objective-dependent infill screening, and geometry-dependent orientation-placement selection. Across the evaluated candidate sets, selected plans achieve up to 53.5% lower maximum Joint-6 jerk and 48.3% lower mean absolute Joint-6 jerk than the least favorable valid candidates, while objective-specific infill screening yields motion-plan completion times up to 40.1% shorter and extrusion paths up to 12.7% shorter than the corresponding least favorable screened patterns.

Full-duplex speech-to-speech (S2S) models provide natural, low-latency conversational interaction and would benefit from the ability to use external tools and complete voice-agent tasks. We propose a frontend-backend architecture where a duplex speech-to-text frontend learns to emit a delegation token and forwards streaming ASR transcripts to a text-based backend LLM for tool calls. Tool-call results from the backend are injected back into the frontend through a lightweight prefill-and-repeat mechanism and then synthesized using streaming TTS to the user. Our approach largely preserves regular duplex turn-taking, interruption handling, and low-latency interaction as it requires minimal modifications to the frontend model. In a single-turn tool-call evaluation, our system achieves 92-97% tool-call recall, competitive tool-call prediction performance, and 81.2% accuracy in rejecting irrelevant calls. When equipped with a larger backend (e.g., Qwen3-235B-A22B), our system achieves competitive results on Full-Duplex-Bench-V3 compared to open and closed source models, and significantly outperforms GPT-realtime-mini and Qwen3-Omni-30B-A3B-Instruct on EVA-Bench. These results demonstrate that backend delegation is an effective and modular approach for combining natural duplex speech interaction with strong agentic tool-call capabilities.

This paper develops a guidance framework for cooperative transportation of a rigid payload by two uncrewed aerial vehicles (UAVs) to stationary and maneuvering landing platforms. A virtual equivalent-agent representation is first introduced to describe the translational motion of the rigidly coupled UAV-payload system, allowing the transportation problem to be formulated in terms of relative range and line-of-sight dynamics with respect to the landing platform. A geometric analysis establishes the terminal feasibility conditions for payload delivery. In particular, an arbitrary prescribed approach angle can be achieved for a stationary platform, whereas successful delivery to a maneuvering platform with zero relative velocity requires terminal velocity and heading angle synchronization and consequently a zero landing angle. Leveraging this framework, a robust fixed-time sliding mode guidance strategy is developed to regulate both relative range and line-of-sight dynamics. A separate link-orientation controller and control allocation scheme is presented to map virtual equivalent agent commands to the individual UAV's control inputs. The proposed strategy guarantees convergence to the desired landing configuration within a uniformly bounded time, independent of initial engagement geometries, while explicitly accommodating uncertainties arising from target maneuvers. Numerical simulations demonstrate accurate delivery under different terminal approach angles and platform maneuvers, while processor-in-the-loop implementation on a Raspberry Pi demonstrates that the guidance algorithm satisfies the real-time computational requirements. Nonetheless, a comparative analysis shows that the proposed framework achieves better tracking accuracy and faster sliding surface convergence while requiring significantly less control energy from each UAV.

Direct preference alignment methods are widely used to align large language models (LLMs) with human preferences because of their computational and memory efficiency. However, likelihood displacement motivates alternative ways to extract information from preference pairs with small likelihood margins. In this paper, we propose and analyze Comparison-based Preference Optimization (ComPO), a zeroth-order alignment method based on comparison oracles. ComPO extracts directional information from these pairs without directly optimizing a differentiable preference loss on them. We establish a convergence guarantee for its basic offline scheme under smoothness, gradient sparsity, and compatibility between the oracle and a latent objective. We further introduce online ComPO, which retains the offline comparison mechanism and uses unlabeled policy generations for reverse-KL control relative to a reference policy. Following the coverage perspective of preference fine-tuning, we establish a performance guarantee for a basic constrained scheme under local coverage and in-distribution pairwise reward accuracy. Experiments on Mistral, Llama, Gemma-2, Qwen3, and Gemma-3 models demonstrate improvements over existing direct alignment methods, including length-controlled win rates, with pair-level diagnostics providing evidence consistent with mitigating likelihood displacement.

Intelligent systems that act in the world require image understanding that is both comprehensive and spatially grounded. Current vision-language models (VLMs) can generate fluent and detailed image captions, but reliably associating them with image pixels remains challenging. Existing methods that combine dense captioning with pixel-level grounding often produce either incomplete descriptions or inaccurate segmentation masks. We study this problem through panoptic grounded captioning, a task that requires a VLM to describe both foreground objects and background regions while grounding each referring phrase with pixel-level masks. We make three contributions. First, we introduce PanoCaps, a human-annotated benchmark constructed from panoptic segmentation datasets. It provides dense captions with near-complete pixel coverage and image-text alignments at the entity level, supporting both training and evaluation. We further propose a phrase-mask matching protocol and a generalized Panoptic Quality (gPQ) metric that jointly evaluates textual and mask agreement. Second, we formulate phrase grounding as selection from a phrase-conditioned pool of mask proposals and introduce PANORAMA, a VLM that conditions a pretrained segmenter on contextualized phrase representations to obtain candidate masks and learns to select those corresponding to each phrase. Training this interface jointly with caption generation enables PANORAMA to produce high-quality masks while allowing each phrase to refer to a single region or multiple instances. Third, PANORAMA achieves the best overall grounding on PanoCaps and matches or exceeds specialized models across several pixel-level grounding tasks. Experiments show that our method produces precise entity-level segmentations while maintaining detailed, mask-consistent captions. Code, data and models are available at https://www.di.ens.fr/willow/research/panorama/.

Scientific code repositories encode decades of human knowledge in executable models, methods, and tools. Yet fragmented toolchains, implicit domain conventions, and specialized correctness criteria make this knowledge difficult to convert into reliable learning experience-a challenge we call the scientific experience bottleneck. We introduce ScienceIDE, infrastructure for turning the world's scientific code into programmable environments for scientific agents. Guided by expert-defined scientific cases and acceptance criteria, agents transform repositories into executable environments that support task generation, execution, and scientific verification. These environments provide a shared foundation for supervised fine-tuning, reinforcement learning, and evaluation. Using verified interaction trajectories, we train PhAI-IDE-72B, PhAI-IDE-9B, and PhAI-IDE-4B. The model family shows gains in held-out scientific-code repair and across selected general-purpose benchmarks in code, reasoning, and knowledge, providing evidence of positive transfer from scientific experience to broader capabilities. ScienceIDE lays the foundation for an integrated workspace for agent learning and scientific practice, making humanity's scientific software a shared substrate for developing scientific intelligence. Code: https://github.com/aitofound/ScienceIDE

Language agents remain brittle in interactive environments, where success requires long-horizon state tracking, valid action execution, and recovery from failed steps. We extend SwiftSage, a dual-process agent that combines a fast action proposer with a slower planner, using two modular cognitive extensions: an Adaptive Memory Module (AMM) for salience-gated episodic storage and trigger-driven retrieval, and a Self-Reflection Module (SRM) for bounded execution-time validation and corrective intervention. Both modules are implemented as feature-flagged extensions over the same execution substrate, enabling controlled ablations on ScienceWorld. Across four configurations---baseline, baseline+AMM, baseline+SRM, and the full system---the full system achieves the best mean final score (64.62), success rate (43.17%), and successful-step efficiency (19.33 steps), while SRM is the strongest standalone contributor. The results suggest that execution-time control is the dominant bottleneck in this setting, while episodic memory becomes most useful once the runtime loop is stabilized.

Emergent coordinated behaviors of AI agents are starting to present critical safety risks. A key phenomenon driving these behaviors is the rapid formation and spread of beliefs about the world, and mechanistic understanding is crucial for collective alignment. To this end, we introduce the Flag Game, a toy model for studying the mechanisms of collective belief formation. Concretely, a hidden country flag defines the ground truth, and each bounded agent directly observes only a private crop but can exchange beliefs and weigh social evidence from peers. Despite its simplicity, the Flag Game reproduces rich collective phenomenology: non-monotonic scaling of performance with population size, accuracy gains from social-awareness prompting and team diversity, and strong effects of organizational structure. In particular, we identify that collective belief collapse at small population sizes turns into collective belief polarization as the population grows. This polarization causes the performance decline at large population sizes, but creates diversity in collective beliefs. Finally, we dissect the mechanisms underlying collective belief collapse and polarization with two complementary approaches. We first introduce social circuit attribution, a technique to predict which agent, and what view, matters most to collective dynamics, and verify its predictions by causal interventions on agents, tracing how agent patching changes collective outcomes. However, the efficacy of causal interventions on agents decreases as the population grows. We therefore develop a statistical mechanical theory for larger populations and verify that it matches the empirical phase diagram. Together, these results take a first step toward mechanistic swarm interpretability, a science of how the properties of individual agents and their communication give rise to emergent collective behavior.

As models scale, reward hacking becomes more frequent, more sophisticated, and more consequential. Does it leave a telltale signature in model representations? This work analyzes how reward hacking is represented internally in frontier open source LLMs, and how those representations can be used to understand and discover the range of hacking behaviors a model displays. In particular, we find that simple difference of means vectors coherently represent reward hacking in Kimi K3, GLM 5.2, and Qwen 3.8 Max across a variety of behaviors in common evaluations. Despite their simplicity, these vectors are both generalizable and interpretable, and we can use them to reliably detect reward hacking. We first evaluate reward hacking in commonly reported benchmarks like DeepSWE and SWE-bench, finding that models reward hack excessively in these environments; GLM 5.2 hacks in 57.2% of rollouts on DeepSWE and in 73% of rollouts on SWE-bench. Catching these requires monitors; LLM monitors are effective, but expensive detectors. We show that DoM vectors are similarly effective but virtually free, catching 3.1% more hacks in Kimi K3 and 7.9% fewer hacks in GLM 5.2 on DeepSWE at a monitor matched false positive rate. DoM vectors run on the chain-of-thought also predict reward hacks in the model's subsequent actions, meaning we can run them online and catch potential hacks before they occur. Finally, we analyze probe-hits that LLM monitors do not catch and discover other undesirable behaviors, as well as show transfer to finding hacks in non-SWE evaluations. Together, these results provide evidence that simple, white-box methods can be used to scalably study and monitor reward hacking behaviors in frontier open source models

Can an attacker turn influence over an artificial intelligence (AI) adviser into a harmful quantum error-correction update? We identify an ambiguity in passive syndrome records that obstructs recovery selection, then show how additional calibration measurements support certified recovery updates under uncertainty and drift. In an odd-distance square toric code with error-free preparation, syndrome measurements, and recovery operations, opposite coherent $X$ rotations produce identical passive syndrome-history distributions. Yet a fixed phase correction can help at one sign and harm at the other. A terminal logical measurement on known encoded calibration states supplies the missing sign information. A separate evaluator accepts an update only when calibration uncertainty and a justified drift bound certify improvement over the current recovery, without assuming that the adviser recommends correctly. In simulated advice attacks, calibration-confidence checks reject harmful proposals while retaining beneficial updates under honest advice. We derive sufficient limits on calibration age that require improvement through deployment. In matched simulations, a validated channel-specific bound retains more beneficial updates than the general bound after accounting for evaluation time, while preventing the tested harmful activations under the stated drift assumption. A separate surface-code experiment includes stochastic circuit faults and noise changing during acquisition. Deterministic controllers achieve at least as many beneficial updates with the same observations. Violating the drift assumption permits harmful acceptance in the toric experiment. The results identify information required for recovery selection, establish conditional guarantees against harmful updates, and quantify the recovery improvements forgone through conservative acceptance.

Large vision-language models have achieved remarkable progress in multi-modal understanding, yet their capabilities in educational settings remain insufficiently evaluated. In AI-assisted language learning, models must interpret artistic imagery, understand its semantic, affective, and cultural content, and reason about visual context to support meaningful interaction. However, existing benchmarks primarily focus on real-world images or domain-specific educational reasoning, providing limited coverage of artistic educational content. To address this gap, we introduce MUSE, a benchmark for evaluating large vision-language models on artistic image understanding in situated educational applications. MUSE decouples image annotation from question generation, enabling diverse tasks with controllable difficulty while reducing annotation effort. It comprises twelve tasks spanning visual perception, semantic and affective interpretation, culture understanding, and compositional reasoning, together with diverse artistic images deliberately curated to center Singaporean and Southeast Asian multicultural contexts alongside Western art traditions, covering multiple themes and difficulty levels. Evaluation of open-source and proprietary models reveals substantial disparities across capability dimensions, particularly in affective interpretation and compositional reasoning. Our analysis further identifies common failure modes and key challenges for developing trustworthy multi-modal models for education. We hope MUSE will serve as a standardized benchmark for advancing multi-modal understanding in situated educational applications.

A game character should not have to reread its entire life before every conversation. For locally deployed language-model characters, however, revising a few memories can invalidate a long reusable prefix. The resulting preparation cost competes with both foreground dialogue and the maintenance of other characters. This matters especially when dialogue feeds game-defined actions and value judgments: a fluent but incorrect account of who owns an item, or whether a transfer has already happened, can corrupt the input to otherwise deterministic rules. We study incremental memory maintenance for long-lived game NPCs in a quantized Qwen hybrid recurrent-attention model. Our runtime removes superseded attention KV entries, computes replacement records at the true sequence tail, and preserves the continuing recurrent state and unchanged KV. Existing local experiments combine multi-update dialogue replays, fixed-input placement ablations, and attention diagnostics. Independent block composition weakens query-conditioned memory selection without a uniform chunk-initial attention collapse. True-tail updates preserve important current-state and historical bindings across eight scripted maintenance rounds; a placement case recovers the full-refill quantity in three reconstructions, while slot-preserving alternatives repeat a double-subtraction error. Attention-distribution proximity alone does not explain these semantic differences. The results motivate treating a character's inference state as a maintained, history-dependent resource, rather than only a disposable encoding of its latest memory text.

In multi-agent environments, coordinating agents to prevent interference and ensure robust individual performance is a critical challenge. Previous research on social laws for multi-agent systems has primarily focused on deterministic, goal-based settings. This paper extends the concept of social laws to stochastic, reward-based environments, proposing a formalism for defining and verifying their robustness under various conditions. We introduce the notion of $α$-robustness, a measure of the guaranteed utility each agent retains while pursuing its optimal single agent policy, assuming all agents obey the social law. We then present an approach for robustness verification of social laws in stochastic settings, based on a reduction to solving a series of Markov decision processes. Empirical evaluations on toy environments illustrate the potential of our framework.

Agent benchmarks are substantially more costly to evaluate than conventional LLM benchmarks. Benchmark compression is therefore a natural solution, yet existing methods primarily model redundancy in task--model final-score distributions, which is important in agentic evaluation. To address this limitation, we analyze large-scale trajectories and identify six complementary process signals that are systematically associated with final agent performance. To disentangle agent performance redundancy from a complete perspective, we propose DualViewEval, an agent benchmark compression method that jointly exploits outcome and process relations to learn an exact-size miniset and predict the full-benchmark scores. Across five agent benchmarks and five representative baselines, DualViewEval achieves the best results in all datasets. With only 20 tasks, it achieves $24\times$--$40\times$ compression on APEX-Agents and BFCL, reducing mean absolute error (MAE) by $14.5\%$--$28.2\%$ over the strongest competitors while improving Kendall's $τ$ by up to $7.2\%$ relative to EssenceBench on SWE-bench Verified. The selected minisets further reveal capability differences among different agents, providing compact and diagnostic feedback for efficient agentic model development.

Existing legal benchmarks cover diverse tasks, while continuous monetary remedies remain comparatively underexplored. We introduce ECtHR-NPD, to the best of our knowledge, the first benchmark for predicting non-pecuniary damage (NPD) awards at the European Court of Human Rights (ECtHR) from case information when no statutory formula or explicit calculation rule determines the amount. ECtHR-NPD contains 14,575 cases with case-level awards in nominal euros, chronological splits, and a protocol separating target construction from model input. We evaluate a battery of methods, including constant predictors, gradient-boosted trees, retrieval methods, fine-tuned encoder language models (LMs), prompted decoder LMs, and knowledge-augmented agents. Our results show that more sophisticated LM and agentic approaches do not consistently outperform the strongest feature-based baseline. All model families struggle to identify zero awards and to calibrate high-award predictions, with further degradation on the Challenging test view, making ECtHR-NPD a challenging testbed for current state-of-the-art open-weight and proprietary LMs.

Health discourse in social media videos often contains densely entangled claims spanning multiple thematic aspects, stances, evidential frames, and rhetorical functions within short conversational spans. Existing approaches largely rely on coarse topic-level, sentiment-based, or stance-oriented representations that do not adequately capture this structure. Our analysis identifies an average of 13.22 atomic claims per minute, motivating richer claim-level discourse representations. We introduce a structured framework for claim-level discourse analysis in dense health narratives. Our framework models discourse through tuples linking atomic claims with thematic aspects, stance, and multidimensional pragmatic discourse attributes. To support this setting, we construct a benchmark spanning four health domains with 1,191 manually annotated claims from 60 videos. Using this framework, we evaluate automated structured discourse analysis under different discourse context settings. Results show that current LLMs achieve strong performance on thematic categorization and stance prediction, but struggle with high-dimensional pragmatic profiling. We also find that different discourse tasks benefit from different forms of contextual reasoning, suggesting that future systems may require task decomposition and specialized inference strategies.

Region-of-attraction (ROA) certificates for polynomial systems become expensive as state dimension and degree grow: direct sum-of-squares (SOS) formulations require combinatorially growing monomial bases. Quadratization represents a polynomial vector field exactly on an invariant manifold of a quadratic system, allowing a quadratic Lyapunov function to certify the ROA. For a fixed lift, stabilizer gains shape the off-manifold extension and transverse dynamics, while representation gauges change the matrix representation but not the vector field. Both affect the spectral-norm certificate, yet prior work fixes the gain by a feasibility heuristic before optimizing the gauge. We formulate optimal dissipative quadratization (ODQ), jointly designing gains and gauges for a fixed monomial lift, reference extension, stabilizer factorization, and Lyapunov weight $Q=I$. Gains lie in a prescribed compact Hurwitz box. At each gain, an exact semidefinite program globally minimizes the spectral-norm bound over the gauge. Residual-aware bounds yield a certified closed Lyapunov sublevel set, accounting for the floating-point Lyapunov residual. Under our stated assumptions, every accumulation point of the idealized outer search is box-Clarke stationary. A finite run returns the best independently verified candidate; global optimality of the gain search is not claimed. On a planar quintic, optimizing the gain increases the certified area by a factor of $2.238$ over a matched zero-gain gauge. Across 16 heterogeneous polynomial systems with stabilizer freedom, ODQ improves on both fixed-gain lifted baselines. All 36 ODQ runs on the relay benchmark complete, and all 27 repeat-level comparisons across nine fully paired cases favor ODQ over an SOS baseline with a fixed quadratic Lyapunov function in both the fixed-direction proxy and construction time. Broader comparisons with direct SOS methods remain mixed.

Adaptive learning systems commonly formulate learning path planning as Exercise-Centric (EC) recommendation, where the next step is inferred from item-level interaction logs. Evaluating goal-oriented guidance additionally requires explicit learner goals and curriculum-scale prerequisites: learners with similar exercise records may need different paths toward their targets. We therefore study Knowledge-Centric (KC) personalized learning path planning, where a planner must reason over learner profiles, mastery states, and prerequisite knowledge structures to decide which textbook, unit, and concept should be studied next. To support this setting, we introduce PersonaPath, a benchmark that pairs 2,000 fine-grained learner personas with a hierarchical knowledge graph of 347 textbooks, 1,751 units, and 4,092 concepts across 77 subjects. We evaluate representative LLMs on PersonaPath. Results show that even the strongest LLM reaches only a 29.5% final pass rate in Basic Education, and that the main bottleneck lies in adaptivity, where no model exceeds 44.7% in tailoring paths to individual learners.

When a large vision-language model misclassifies a harmful meme, the failure may reflect missing internal evidence or an inability to route represented evidence to its output. We distinguish these cases in Gemma-3 and Qwen3.5 using sparse autoencoders, role-conditioned probes, causal interventions, and recovery experiments across six harmful content benchmarks, with additional Spanish and Hindi-English code-mixed evaluations. Sparse readouts outperform native prediction on all six primary binary tasks: Qwen averages $0.740$ versus $0.432$ native macro-F1, while residual reconstruction reaches $0.486$, whereas Gemma improves from $0.532$ to $0.714$. These differences reflect supervised accessibility rather than a pre-existing, native decision rule, and the most influential token role depends on the task. Under the evaluated score scales, Qwen silent-feature ablation is $24-63$ times more probe-sensitive, whereas routed-feature patching on literal yes/no tasks is $16-140$ times more output-sensitive. Calibration-only routing recovers $93.3$% of the mean gap, and probe-distilled LoRA improves native predictions, although shared multi-task adaptation causes negative transfer. A case study of Gemma-3-12B on Facebook Hateful Memes finds a distributed rank-32 image-prompt interaction, reaching $0.756$ versus $0.685$ native macro-F1. Robustness controls show that the signal extends beyond English, is not explained solely by accompanying OCR, and depends on paired visual evidence. Thus, routing, rather than representation alone, is a recurring bottleneck in harmful meme classification.

Two-stage stochastic Mixed-Integer Linear Programs are a canonical modeling tool to optimize power system operations under uncertainty, yet their extensive-form counterparts scale linearly with the number of scenarios and quickly become computationally prohibitive under day-ahead time constraints. We propose an Input Convex Neural Network architecture to learn a convex surrogate of the second-stage value function, enabling fast first-stage optimization while preserving convexity by construction. We couple the surrogate with a Neural-Benders correction loop that refines the first-stage solution a posteriori, recovering the exact optimum whether the network overestimates or underestimates the recourse cost, and certifying solution quality independently of surrogate accuracy. We evaluate the method on IEEE Stochastic Unit Commitment benchmarks (case9-case118) under both continuous and integer recourse. The proposed approach achieves solutions with zero optimality gap on the $K$-scenario instance across all test cases and scenario dimensions $K \in \{10, 50, 100\}$, with speedups up to $214\times$ over the Extensive Form in the integer setting. Solve times are stable across scenario realizations and compatible with day-ahead time windows on the tested benchmark instances.

The O-RAN control plane is becoming agentic: autonomous AI agents, deployed as rApps by different vendors, independently close control loops over shared radio resources. We demonstrate on a live O-RAN system that this independence is unsafe. Two agents with individually correct objectives, one protecting a latency SLA and one maximizing utilization for energy efficiency, jointly drive recurring opposing excursions of the shared resource partition that neither produces alone. Existing conflict-mitigation mechanisms presume a statically known application population and cannot govern agents whose behavior emerges at run time. We present AURA, a lightweight arbitration layer that admits agent actions only when they satisfy feasibility invariants, per-variable dwell times, and a deadband, and we prove the arbitrated system converges to a feasible operating point. Implemented on an OpenAirInterface (OAI) testbed with measured one-way latency and throughput, AURA reduces recurring shared-state excursions by more than an order of magnitude (from 8.4 to 0.4 PRB amplitude) and virtually eliminates cross-slice throughput starvation (from 40-55% to 0.3%), while leaving the protected slice's own latency compliance unchanged, a trade-off the convergence guarantee makes explicit.

A central challenge across control and learning is the design of mechanisms regulating the interactions between humans and autonomous agents. Inspired by the free energy principle from computational neuroscience, we introduce a control-theoretical framework to integrate human preferences online into an agent policy. We turn the framework into an open control architecture and validate our approach using a human-in-the-loop experimental testbed involving a rover navigating via onboard sensing. The human, remotely located and equipped with virtual reality headsets, shares the same sensory information as the rover. Human preferences are provided to the rover via gestures which introduce both cooperative and competitive interactions between the agent goal and the preferences. The experiments show that interactions are regulated, validating the proposed approach.

Longitudinal electronic health records (EHRs) capture years of patient history across notes, codes, labs, and procedures, and contain evidence needed to reason about likely clinical outcomes. However, comprehensive clinician review of these records is impractical, and LLM-based processing is costly and often unreliable, missing some relevant observations while hallucinating others. We therefore propose EviGen, a three-layer framework for verifiable clinical rationale generation that addresses these challenges. The first layer is a patient-conditioned retriever that uses learnable queries to find evidence predictive of, not just textually relevant to, a clinical outcome and ranks it by prediction attribution scores. The second layer is an LLM generator that consumes this ranked evidence as a scaffold to produce a clinical rationale grounded in the retrieved spans. The third layer is a process-supervised verifier that checks the generated rationale at the reasoning-step level, flagging unreliable claims. Across three medical prediction datasets, EviGen improves prediction performance and rationale faithfulness over full-context LLM and RAG baselines, and is preferred by clinical reviewers in a usability evaluation.

The rapid proliferation of grid-edge distributed energy resources has significantly increased the operational complexity of modern power systems. Consequently, conventional computational techniques face growing scalability and computational-efficiency challenges in addressing large-scale optimization and control, uncertainty management, nonlinear dynamics, and combinatorial decision-making in smart grid operations. Quantum computing has therefore emerged as a promising computational paradigm that can complement classical methods in addressing selected computationally intensive problems. In this context, this paper presents a comprehensive structured review of quantum computing applications in smart grid operations. Following a transparent keyword-based literature search, the paper classifies and assesses existing studies across monitoring and estimation, system planning, operation and control, security, reliability and resilience, stability assessment, data-driven intelligence, and digital twin technologies. The paper also describes fundamental quantum-computing concepts and key algorithms, highlighting their relevance for power system applications. Furthermore, it reviews the current state of quantum hardware, software frameworks, simulators, cloud services, and emerging hardware-agnostic ecosystems that support cross-platform application development and deployment. The reviewed studies are examined by implementation environment, benchmarking practices, application scale, and evidence of computational advantage. Finally, the principal challenges associated with practical implementation are discussed, and future research directions are outlined. This paper provides a consolidated and evidence-calibrated perspective on the current state and future potential of quantum computing for next-generation smart grid operations.

Multimodal coding agents are expected to turn visual inputs into usable artifacts, and they act through a harness, the layer of tools, context management, and execution environment around the model. Existing evaluations often isolate short tool calls, API traces, or screenshot resemblance, and a low score under these proxies cannot say whether the model saw poorly, planned poorly, or was failed by its harness. We study scientific overview figure reconstruction, an agent task in which a source image must become an editable PowerPoint slide that preserves text, topology, layout, and native document structure. We introduce ReFigBench, a benchmark and evaluation framework built on 1,000 real overview figures retrieved from arXiv papers with full provenance. Coding agents from four model families reconstruct every figure under two workflows, direct code generation and a specialized PPTX workflow, and the strongest model runs inside two commercial harnesses, yielding ten configurations. Evaluation combines deterministic artifact checks, repeated automated scoring by judges from two model families, and blinded human comparisons. Perception remains a bottleneck that iterative rendering only partly repays. Whether workflow effort converts into quality depends on the model together with its harness, since the same model gains from the specialized workflow inside one harness and loses inside the other, and the harness shifts scores even under an identical direct prompt. The specialized workflow erases native connectors in every configuration, human judges still prefer its renderings in most matchups, and even the strongest agent falls short of the rubric ceiling. These results expose the tension between fidelity and editability as the central challenge for practical multimodal document agents.

Agentic workflows now make consequential decisions in regulated settings, and the governance placed around them is almost entirely step-scoped: input-output classifiers, per turn rails, and span-level evaluators. The policies organizations actually hold, such as referral thresholds, authority limits, and review requirements, are properties of the whole execution rather than of any one step. This mismatch admits a failure mode we call a Compositional Policy Violation (CPV): every individual step passes its own check while the composed execution violates the governing policy. A predicate over a single step cannot evaluate a property that step does not determine, so no improvement in the accuracy of the step-scoped monitors detects this class. We define CPVs as the failure of step-level compliance to compose, and present a taxonomy of four types: Authority Creep, Threshold Laundering, Cumulative Sum Violation, and Context Collapse. We show that the correct repair for each class is dictated by where the guarded quantity mutates. We then introduce a provenance-aware runtime architecture that evaluates policies over complete execution traces, recomputing guarded quantities from raw provenance rather than the pipeline's derived representation.

Dual-comb technology has emerged as an essential tool for high-precision spectroscopy, real-time ranging, and high-sensitivity sensing. Integrating dual-comb sources into a single microresonator would substantially reduce pump power, footprint, system complexity, and cost, yet this remains a significant challenge. Here, we demonstrate, for the first time, integrated dual-comb generation in a single thin-film lithium niobate (TFLN) microring, under single continuous-wave laser pumping. Rather than regarding TFLN's strong Raman nonlinearity as detrimental, as conventionally viewed, we harness it constructively. By engineering the dispersion of TFLN microrings, we leverage the fundamental and first-order transverse-electric mode families with loaded Q factors exceeding 5X10^6, comparable repetition rates, and suitable dispersion profiles, and bridge them through stimulated Raman scattering (SRS) processes. Pumping a first-order mode at 1551.28 nm initially excites both Stokes and anti-Stokes SRS in the fundamental mode family at low thresholds, and subsequently produces two independent, spectrally separated combs at a pump power of 320 mW via direct Kerr and Raman-assisted Kerr effects, respectively. The two combs span broad bandwidths, exhibit repetition rates of ~102 GHz with a slight difference of ~624 MHz, and do not merge spectrally. The broadest spectrum spans 654 nm, and the Raman-Kerr comb has a 3-dB bandwidth exceeding 29 nm. Further characterization confirms that the comb lines exhibit low phase noise, with an intrinsic linewidth of 410 Hz. This work establishes a robust pathway for on-chip dual-comb generation in a single-laser pumped microring, significantly advancing dual-comb systems toward simplified architectures, enhanced robustness, and scalable integration, while accelerating their practical deployment.

Safe motion planning in automated driving requires anticipating evolving traffic risks and deciding when to revise the current planned trajectory. We introduce RiskWorld, a risk-aware world modeling framework for shared occupancy forecasting and selective trajectory replacement. Spatial risk fields and temporal actor context are fused with visual bird's-eye-view features. Flow-guided evolution transports occupancy and scene features, while signed residuals correct occupancy after transport. One forecast is generated per planning step and reused across candidates. Each candidate is compared with a current-state persistence reference, yielding a nonnegative collision-score correction. The trajectory selected by current-world evaluation serves as the planning anchor and is replaced only when additional predicted risk triggers intervention and an alternative satisfies component-wise constraints on predicted risk and trajectory error. Candidate geometries remain unchanged. We evaluate RiskWorld for open-loop planning on nuScenes using camera features, annotation-derived current and historical actor states, and dataset-provided map context. RiskWorld achieves the lowest collision rate at a long evaluation horizon of 3 s, and the second-best average L2 error among various state-of-the-art baselines, while running at 11.5 FPS on a single NVIDIA RTX 4090 with 90.81 M parameters. Within-setting ablations show that RiskWorld achieves lower collision rates than the current-state rescoring baseline, while forecast reuse enables additional candidates to be evaluated at low marginal computational cost.

Agentic browsers can execute security-sensitive actions under a user's authenticated session, making indirect prompt injection and deceptive confirmation interfaces a direct threat to action integrity. Existing human-in-the-loop (HITL) safeguards are insufficient when the approval prompt itself can be influenced by untrusted page content or model-generated text. We present the \emph{Verifiable Action Card} (VAC), an architectural defence that reconstructs approval information from the ground-truth pending browser action and trusted intent provenance, renders it out-of-band in the trusted browser chrome, and binds approval to the exact action re-verified at dispatch. VAC combines provenance fencing, a ground-truth action descriptor, default-deny confirmation, provenance-aware risk gating, and execution binding. We implement VAC in a complete agentic browser and evaluate it on a 24-scenario benchmark covering confused-deputy attacks, Lies-in-the-Loop dialog forging, indirect prompt injection, adaptive action substitution, provenance evasion, and legitimate tasks. Across the evaluated LLMs, attack success without VAC ranges from $68\%$ to $100\%$, whereas VAC reduces attack success to $0\%$ on every model, with $78\%$ legitimate-task completion and a $0\%$ false-block rate. These results show that grounding approval in the action that will actually execute provides architectural protection against security failures that prompt-level defences and conventional HITL confirmation cannot reliably prevent.

AI applications have shifted from single, monolithic foundation models (FM) to compound agentic systems. Yet today's stacks remain fragmented: even as protocols (e.g., MCP, A2A) ease tool/agent connectivity, each framework embeds an implicit runtime for state, memory, budgets, and guardrails, making behavior non-portable and governance brittle. It mirrors computing before operating systems, when every program re-implemented basic services. This position paper argues that the field now needs a Foundation Model Operating System (FMOS) -- a system layer that virtualizes FM interactions analogous to how virtual machines abstract physical hardware, giving applications the illusion of dedicated, trustworthy FM instances with effectively unbounded capabilities. Internally, the FMOS orchestrates knowledge across memory tiers, model selection and resource allocation, and verification and policy enforcement. Like the human brain switching between fast intuition and slow deliberation, the FMOS learns when to intervene and when to let inference proceed directly and continuously adapting its policies based on operational experience.

Mechanical memristors are systems whose dissipative response depends on the history of previous loading through an evolving internal state. History-dependent forces and dissipation occur in a wide range of materials and devices, including viscoelastic polymers, shape-memory materials, piezoelectrics, granular media and field-responsive fluids. Determining which of these responses admits a mechanical-memristor representation requires a constitutive test, as well as a comparison of scales. In this work, a fractional-order mechanical memristor model is developed and cast into a nondimensional form to identify the governing parameters controlling memory-dependent dissipation. The formulation leads to a set of dimensionless groups that characterise dissipation magnitude, memory-state scale and memory transfer. These quantities are combined into an effective mechanical memristance screening index \(\Mh = βγ|\mathcal H_α(Ω)|\), which provides a conditional measure of local damping modulation at matched response amplitude, constitutive slope and reference scales. Illustrative parameter scenarios are then constructed for material classes including shape-memory polymers, shape-memory alloys, hydrogels, nanocellulose, lignin-rich materials, natural fibres, piezoelectric polymers, piezoelectric ceramics, electrorheological fluids, magnetorheological fluids and granular dampers. The framework establishes a common basis for comparing memory-dependent damping within the adopted constitutive description and identifies the calibration required for its application to candidate materials and devices

The elements of a reconfigurable intelligent surface (RIS) are commonly modeled either as frequency-selective time-invariant reflectors or as instantaneous time-varying reflection coefficients. In practice, however, time-modulated metasurfaces exhibit both frequency selectivity and periodic time variation. We develop a physically consistent linear periodically time-varying (LPTV) model that jointly captures these effects and characterizes their impact on wideband orthogonal frequency-division multiplexing (OFDM) communications. From a canonical equivalent circuit of a generic RIS element, we derive a single-resonance model whose physically meaningful parameters determine both the frequency-selective reflection coefficient and the effective impulse-response duration, i.e., the finite memory of the element. The periodically switched dispersive responses are then represented through harmonic transfer functions, leading to a closed-form per-subcarrier OFDM input-output relation. The resulting coupling is generally non-diagonal: each received subcarrier collects contributions from multiple transmitted subcarriers through the RIS harmonics, each weighted by the element response at the corresponding absolute input frequency. We further derive a generalized cyclic-prefix (CP) condition requiring the guard interval to accommodate both the propagation-channel delay spread and the RIS memory. Under this condition, intersymbol interference and out-of-grid spectral leakage are suppressed, while deterministic on-grid harmonic coupling remains. Full-wave simulations of an OpenRIS unit cell designed for a 5G NR channel validate the proposed resonant model and reveal appreciable in-band dispersion despite nearly ideal binary phase switching. Simulations over 3GPP tapped-delay-line (TDL) channels confirm the generalized CP condition and show its relevance for high-quality-factor RIS elements.

Battery Energy Storage Systems (BESS) operators face significant challenges when participating in multiple electricity markets due to the complex coupling of price volatility and stochastic reserve activation. Traditional deterministic dispatch models neglect the "tail risks" associated with extreme market realizations, potentially leading to technical infeasibility or severe economic losses. This paper proposes a risk-aware stochastic optimization framework for the co-optimization of BESS participation in the Day-Ahead (DA) energy market and the manual Frequency Restoration Reserve (mFRR) market. The model explicitly captures multi-dimensional uncertainties by utilizing non-parametric Kernel Density Estimation (KDE) to generate joint price scenarios that preserve the empirical characteristics of European balancing markets. A two-stage stochastic programming approach is employed.To manage the financial exposure to high-impact price events, the Conditional Value-at-Risk (CVaR) metric is integrated into a Mean-CVaR objective function. This allows decision-makers to tune their risk-aversion levels and identify an efficient frontier between profitability and robustness. Simulation results demonstrate that the proposed joint CVaR approach significantly enhances revenue stability compared to deterministic benchmarks.

As telecommunication networks evolve toward autonomous 5G-Advanced and 6G operations, agentic artificial intelligence (AI) workflows, where large language models (LLMs) execute multi-step reasoning, invoke diagnostic tools, retrieve domain knowledge, and coordinate across agent teams, are increasingly embedded across the edge-cloud continuum. While the biological brain accomplishes complex cognition on an exceptionally modest metabolic power budget of approximately 20W contemporary LLMs are profoundly energy- and memory-intensive, making sustainable lifecycle orchestration a critical operational priority. However, existing AI lifecycle metrics evaluate only isolated, single-model inferences or overlook multi-agent execution graphs entirely. Consequently, network operators lack foundational models to determine whether distributed agent communication incurs meaningful energy costs and where across edge-cloud tiers agent teams should physically reside. To address this gap, we introduce agentic-eCAL, generalizing the Energy Cost of AI Lifecycle (eCAL) metric to directed multi-agent workflows by coupling a closed-form two-rate single-call energy model (compute-bound prefill and memory-bound decode) with 7-layer OSI data transport. Grounded in hundreds of GPU benchmark configurations on NVIDIA A100 and H100, 16 open-weight models and 8 orchestration topologies, we validate components of the metric and study workflow placement implications. Our findings demonstrate that inter-agent text transport incurs 0.25% of workflow energy across 5G RAN, metro, and optical links. Therefore in edge-cloud agent placement the dominant energy cost of distribution is often not the transmission of inter-agent text itself, but the additional inference and context processing induced by that communication.

Traditional graph rigidity theory enforces structural constraints in the position space, effectively restricting a multi-agent formation to a static geometric shape. In this paper, we introduce a fundamental paradigm shift by applying rigidity constraints directly to the agents' continuous-time velocity space. We propose the concept of generalized velocity rigidity, demonstrating that the macroscopic physical motion patterns of a multi-agent network are entirely dictated by the underlying static graph topology. By strictly confining the network's acceleration profile to the null space of the velocity rigidity matrix, we mathematically map the trivial motions of this null space into exact physical trajectories. Specifically, we prove that translational, rotational, and scaling trivial motions seamlessly integrate into cohesive curved flocking, synchronized helical and circular $\mathrm{SO}(d)$ orbiting, and exponential spatial homothety, respectively. Furthermore, we analyze velocity-space singularities, showing that velocity consensus induces a valid topological contraction rather than a structural failure. Finally, we provide comprehensive numerical simulations to validate these theoretical mappings, demonstrating that complex macroscopic maneuvers can be orchestrated purely through algebraic constraints in the velocity space, unlocking unprecedented spatial shape flexibility for multi-agent swarms.

Aerial grasping is a remarkable capability exhibited by predatory birds, allowing them to capture prey through highly coordinated maneuvers in flight. Inspired by this capability, researchers have developed various formulations to reproduce such maneuvers through trajectory optimization. However, two limitations remain in practice. First, the resulting optimization problem is highly nonconvex and sensitive to initialization, making high-quality solutions difficult to obtain under a limited computational budget. Second, prescribed numerical objectives are human-designed abstractions that describe successful grasping through a limited set of mathematically tractable quantities and may not fully capture what determines task success. We investigate how learning can address these limitations within an analytical planner. Accordingly, a trajectory prior is first learned from optimized motions and then evolved through a CEM-based process that evaluates sampled initializations with the deployed optimizer and retains favorable ones as new supervision. An Execution-Aware Critic learns from contact, lift, and completion outcomes to assess whether the optimized trajectories are likely to succeed in physical execution. Its frozen energy can further serve as a differentiable grasping cost, allowing execution data to directly shape trajectory generation. Simulation and real-world experiments demonstrate improved optimization reliability, trajectory consistency, and grasping performance.

An agentic framework is proposed for autonomous site-specific base station (BS) deployment in wireless network planning. In contrast to conventional approaches that rely on manual site surveys or extensive ray-tracing (RT) simulations with significant human intervention, the proposed framework autonomously explores and optimizes BS deployment under a limited RT evaluation budget, enabling resource-efficient network planning. To this end, a continuous, geometry-grounded deployment action space is first constructed from three-dimensional (3D) wireless digital twins. Within this action space, an agent team operates through a stateful perception--reasoning--reflection loop. Specifically, a Placement Agent first generates candidate BS deployments in two complementary modes: an experience-guided mode that refines promising solutions, and an exploration mode that avoids getting stuck in local optima. After the candidate deployments are evaluated through RT, a Reflection Agent interprets the RT results together with the scene geometry, identifies performance-limiting factors such as blockage, overlapping coverage, and uncovered areas, and converts these diagnoses into guidance for subsequent deployment. Through this iterative process, site-specific experience is accumulated and deployment plans are optimized without human intervention. Numerical results in two realistic urban scenarios show that: 1) the proposed approach substantially outperforms heuristic, learning-based, and large language model (LLM)-assisted methods; 2) it achieves highly competitive coverage against the optimal solution while requiring substantially fewer transmitter-level RT evaluations; 3) site-specific reflection effectively turns raw RT feedback into refinement guidance, whereas the dual-mode mechanism preserves diversity and facilitates escape from local optima.

Contemporary memory technologies are increasingly constrained by the fundamental trilemma of storage capacity, access latency, and power consumption. Among the emerging technologies, spin-orbit torque magnetic random-access memory (SOT-MRAM) shows promise to circumvent these challenges, owing to its fast switching dynamics and high endurance. However, the application of SOT-MRAM is hindered by the relatively low write and read efficiencies, resulting in a large bitcell area and an insufficient sensing margin. Meanwhile, the involvement of an ultrathin spin-source channel, typically within a few nanometers, imposes technological challenges for mass production. Here, we resolve these issues on a 300-mm wafer platform by exploiting the emerging orbital degree of freedom and the resultant orbital torque (OT) from the relatively thick Ti/W bilayer. In particular, OT memory nanodevices exhibit a giant tunnel magnetoresistance (TMR) of 182%, nanosecond-scale response, 1012 endurance, together with an enhanced switching efficiency (E_b/I_c), which consequently enables an ultra-low write energy of less than 0.1 pJ/bit. Our findings demonstrate that orbital angular momentum can be implemented for building energy-efficient MRAM devices, offering a practical pathway towards low-latency memory that is demanded for high-performance computing and AI applications.

Computer-use agents that operate through screenshots and simulated actions are advancing rapidly, yet their evaluation remains anchored to general desktop and web tasks. Enterprise Resource Planning (ERP) systems run the finance, procurement, inventory, and customer operations of organizations worldwide, and pose distinct challenges for computer-use agents: dense interfaces, coordinated multi-step interactions, and errors that alter persistent business records rather than surfacing on screen. Existing enterprise benchmarks rely on proprietary platforms or on simulated approximations of such software. We introduce ERPBench, a benchmark that evaluates screenshot-only agents on a live and reproducible ERP system and scores each task against ground-truth values in its database. Beyond the benchmark, we present a production-grade harness that gates agent actions behind human approval for safe deployment, which ERPBench runs autonomously. Evaluating six closed and open-source agents, we demonstrate that strong general GUI performance does not transfer to enterprise reliability. Even when an agent reaches the right form and saves it, the stored record is often wrong: some agents save in up to 85% of runs but write the correct value in as few as 3%. We further characterize failure modes specific to enterprise workflows.

Heterogeneous cooperative perception (CP) enables connected vehicles with diverse sensor setups to share spatial awareness via compact feature maps, where receivers reconcile these maps using learned translation modules for fusion and inference. Prior attacks against CP in a homogeneous setting reveal that the data exchange introduces a critical attack surface: a single malicious agent can transmit crafted features that erase real objects from a neighbor's fused scene. Yet, it is widely hypothesized that heterogeneity naturally defends against these attacks, as the attacker lacks knowledge of the victim's detector and the translation module scrambles adversarial gradients. We demonstrate that this protection is largely an illusion. Using a matched-objective harness to standardize the perturbation budget, objective, and forward path, we show that properly tuned iterative attacks close or reverse the apparent robustness gap. However, these optimization-based attacks require ground-truth labels and iterative backpropagation, meaning they do not represent a practical field threat running in real-time. To bridge this gap, we introduce HetPoison, a learned generator that crafts a removal perturbation in a single, label-free forward pass. HetPoison transfers across major heterogeneous designs without requiring access to the victim's detector, matching or exceeding the effectiveness of expensive optimizer-based attacks. Since heterogeneity itself is not a defense, we propose HetShield, a lightweight trust layer that validates the spatiotemporal consistency across features, recovering 83--95% of the accuracy degraded by attacks, outperforming prior art.

Quantum machine learning offers a compelling representational perspective: angle-encoded states inhabit Hilbert spaces in which periodic similarities and interactions can be expressed naturally. Realizing this perspective for visual recognition remains difficult, however, because present quantum neural networks are constrained by limited qubit counts, costly circuit simulation and measurement, noise, and unstable optimization on noisy intermediate-scale quantum devices. We investigate whether useful structural ideas from quantum models can instead be realized as scalable classical Transformer operations. We introduce QiT, a Quantum-inspired Transformer for vision tasks with three components: (i) angle-inspired encoding that maps image tokens to learned trigonometric Hilbert-space features analogous to quantum rotation-based state encoding; (ii) self-attention over these periodic features, inducing a classical cosine kernel approximated to quantum fidelity kernels; and (iii) gated multiplicative emulation, a trainable classical surrogate for interaction terms found in variational circuits. All components are differentiable tensor operations, so QiT claims neither quantum computation nor quantum speedup and retains the $\mathcal{O}(N^2D)$ attention complexity of a standard Vision Transformer. Across image-classification benchmarks, QiT is competitive with a matched classical Transformer while avoiding the severe runtime cost observed for a small simulated quantum Transformer. QiT-B reaches 78.3\% ImageNet-1K top-1 accuracy with 45.7M parameters and 11.5 GFLOPs. These results position QiT as a scalable baseline for isolating and evaluating quantum-motivated inductive biases in visual recognition.

We combine the Briegel--Englert damping basis of a lossy cavity mode with the permutation-invariant Liouville basis of \(N\) identical, incoherently pumped two-level emitters. The resulting Liouvillian is exactly block tridiagonal in a single radial field index, with atomic blocks of dimension \(\binom{N+3}{3}\). The stationary state follows from a matrix continued fraction whose asymptotic closure is a linear matrix equation that we prove to be uniquely solvable throughout the dissipative regime \(A>0\), and the same construction in the coherence-one sector yields the emission spectrum. The method is validated against direct diagonalization, weak-pump perturbation theory, and closed-system spectra, reaches \(N=16\) in tens of seconds in the few-quanta regime, and has a quantified precision limit at large photon number. We use it to derive an exact two-atom photon-balance identity that isolates the cavity-induced pair coherence and relates it to a singlet--triplet population difference; to establish an exact pump-order hierarchy of connected correlations for a zero-temperature cavity and a parity property of the minimal collective-spin sector; and to benchmark the second-order cumulant closure used in few-emitter laser theory, which we find to be semi-quantitative for the photon number but to predict the wrong sign of the inter-emitter coherence above inversion. In a good-cavity regime the exact solution shows near-Poissonian statistics and strong line narrowing for two and three emitters.

Integrated deep reinforcement learning (DRL) and model predictive control (MPC) methods are increasingly used to control autonomous systems by combining their complementary capabilities. DRL learns control policies through interaction with the environment. MPC uses a system model to optimize control inputs while accounting for constraints. In DRL-MPC frameworks with shared control authority, both the DRL agent and the MPC controller each determine part of the control inputs. However, common learning formulations treat MPC as part of the environment and therefore do not explicitly account for MPC's contribution to control or its interaction with the DRL agent. This paper proposes a novel composite-gradient learning (CGL) method that integrates the MPC controller into the learning process by representing the DRL and MPC control inputs as a joint action and accounting for their interaction when updating the DRL agent during training. CGL is evaluated on two multi-class freeway traffic networks with different strengths of interaction between the DRL and MPC control inputs and it is compared with alternative methods that treat MPC as part of the environment or that only partially incorporate MPC into learning. The results show that CGL offers limited benefit under weak interaction, but learns higher-performing control policies than the alternative methods in a subset of training runs under strong interaction, although the average control-performance gains remain modest.

An agentic society is a collection of AI agents that coordinate autonomously across trust boundaries, on behalf of different principals whose objectives may only partially align. We show experimentally that in agentic societies even honest, competent agents often fail to reach satisfactory outcomes with existing harnesses and messaging primitives, and that faulty or malicious agents can stall collaboration, influence outcomes, and pursue other harmful goals by exploiting vulnerabilities in communication (``speech''). We argue that agentic societies need a \emph{social harness} for inter-agent interactions, in addition to each agent's \emph{personal harness}, which manages its private context and communication with its principal. We propose a layered architecture for social harnesses which (i) prevents classes of failures outright, (ii) enables agents to detect invalid messages at runtime, and (iii) supports post-facto investigation and consequences, and highlight directions for future research to realize these capabilities.

$\text{SiGeSn}$ alloys are promising silicon-compatible semiconductors for monolithic infrared photonics. However, their metastable nature limits the thermal budgets available for post-growth device processing, and the mechanisms governing their thermal degradation remain unresolved. Here, we investigate the thermal stability of $\text{Si}_{0.08}\text{Ge}_{0.83}\text{Sn}_{0.04}$ alloys using in situ spectroscopic ellipsometry (SE) during isothermal annealing at 550 °C. We show that adding an ultrathin oxide cap kinetically suppresses Sn exchange with the free surface while leaving bulk diffusion pathways largely unaffected. Uncapped films undergo phase separation after 50 min, accompanied by void formation, a 60% thickness reduction, and a 400 meV blueshift of the $E_{2}$ critical point (CP) transition, consistent with substitutional Sn depletion from the probed volume through surface segregation. In contrast, oxide-capped films exhibit a small compositional change (<1 at.% Sn) and optical shift (<20 meV) over the same period, with suppressed void formation, strain relaxation, and alloy decomposition. This surface-kinetic control additionally yields a 25-fold reduction in contact resistivity relative to annealed uncapped alloys. These results identify surface Sn transport as the dominant degradation pathway in SiGeSn and demonstrate that an ultrathin oxide cap extends the thermal stability of metastable group-IV alloys, providing a practical route toward their integration into advanced silicon photonic and electronic platforms.

Silicon nitride photonic integrated circuits for the visible band are conventionally built with a buried oxide overcladding and singulated with wafer-scale tooling, constraints that preclude evanescent access to the guided mode for externally integrated emitters. We report a PECVD grown $Si_3N_4$ platform designed around an air-clad waveguide whose evanescent field remains accessible along the full device length. Two process elements make this geometry practical at chip scale. Fixed-beam moving-stage electron-beam lithography writes 500 nm single-mode waveguides as one continuous exposure across the 5 mm chip, removing write-field stitching which, given the $σ^{2}/d^{4}$ scaling of sidewall scattering in this high-confinement geometry at 635 nm, would otherwise dominate the loss budget. Chip singulation is performed by pen-type diamond scribing along lithographically patterned markers registered to in-plane direction, cleaving the Si(100) substrate to yield end-facets within $2^\circ$ of normal at $80 \%$ yield. Structural characterization by scanning electron microscopy confirms stitch-free waveguide geometry and undamaged, near-vertical scribed facets; light is coupled end-fire into fabricated devices and guided to a microring with evanescent bus-to-ring coupling confirmed by scattering imaging, and a sidewall-roughness-dependent scattering-loss model indicates that loss remains low in the roughness regime consistent with the observed facet and sidewall quality. Building on the intrinsic emitter-resonator coupling demonstrated in, this platform extends monolithic $Si_3N_4$ photonics toward scalable visible-to-near-infrared quantum and classical circuits.

Multi-agent systems split a task across a tree of agents and justify the split with folklore: smaller contexts, cleaner separation, parallelism. We ask what the split does to how much of what the leaves discover reaches the root. Model a decomposition as a tree in which an agent handed $b$ items keeps any one with probability $r(b)$. If $r(b)=1/b$, every tree delivers exactly one finding, for every task size and every shape; we verify this to $2.4 \times 10^{-15}$ on 20,000 random irregular trees. If $r(b)=Cb^{-δ}$, a depth-$k$ tree over $N$ findings yields $C^k N^{1-δ}$: task size and architecture separate, and architecture contributes only $C \le 1$ per level, so flat is optimal for yield and no arrangement of agents escapes the exponent $δ$. On 600 production deep-research traces $δ= 0.34$ [0.30, 0.38], by three identifications that do not share a failure mode. At a hop where item boundaries come from the tool rather than a text heuristic, and where $b=1$ occurs 550 times, $C = 0.571$ [0.527, 0.615] is observed rather than extrapolated, over 16,082 hops. A tier also costs alignment: on 1,012 annotated multi-agent traces one brief in sixteen goes off-target, giving $μ= 0.939$ and a per-tier penalty $Cμ= 0.536$. Depth is bought on two other axes. The root context is the only state that persists and the only one that cannot cheaply forget, and depth cuts its exposure from $N$ items to $N^{1/k}$. Depth is also cheaper: production flat agents bill as $N^{1.39}$, not the $N^2$ an append-only context predicts, and at equal spend two tiers overtake flat at 403 findings. Across every parameter we measured the model says 0.7% to 11.3% of production sessions are worth delegating, against 7.8% that do. A hazard model on 743,819 production tool calls finds that delegation does not respond to a filling context and is instead an opening move.

Hybrid dynamical systems provide a powerful modeling framework for robotic systems, particularly in contact-rich environments. However, ensuring safety and performance in such systems remains challenging due to the intricate coupling between continuous dynamics and discrete mode transitions. In this work, we extend classical Hamilton-Jacobi (HJ) reachability analysis, a formal verification method for continuous-time nonlinear systems, to hybrid dynamical systems. Our framework characterizes safe sets for hybrid systems through a generalized value function defined over both discrete and continuous states while accounting for control constraints and model uncertainty. We additionally provide a numerical algorithm to compute this value function. Building on these safe sets, we propose two different mechanisms to integrate performance objectives. First, we introduce a hybrid least-restrictive safety filter that intervenes on both the discrete and continuous components of a nominal controller only when necessary to avoid unsafe states, thereby preserving nominal behavior whenever possible. Second, we formulate and compute hybrid backward reach-avoid tubes, enabling the simultaneous enforcement of safety and goal-reaching behavior, an extension not previously addressed within hybrid HJ reachability. This enables the synthesis of continuous and discrete control policies that guarantee both safety and task completion. We validate our framework through simulation studies and real-world experiments on a quadrupedal robot, demonstrating its effectiveness in hybrid mode planning and safety-critical applications.

Loading multiple quantum states in parallel into a quantum machine learning (QML) model can unlock learning tasks where key information resides in the \emph{relations} between states rather than in individual states. We introduce an adaptive relational learning framework for such multi-instance quantum data that accesses pairwise and higher-order relations. Our model combines global measurements via SWAP or CYCLE tests for evaluating an $n$-state Bargmann invariant with shallow trainable transformations applied locally to each input state. We demonstrate the approach for continuous-variable (CV) photonic systems, which naturally provide access to quantum data and necessary computing operations. We solve tasks involving hidden relationship detection, geometric phase classification, and sensing in the presence of an unknown shared nuisance interaction. We benchmark the adaptive model against a non-adaptive ``measure-first'' approach based on continuous-variable classical shadows, and show that the cost of shadow estimation grows rapidly with $n$, while our model avoids this dependence. Already for $n=2$, we achieve perfect test accuracy $A=1.0$ with $500$ inference shots, improving average test accuracy over the shadow-based method by $ΔA=0.15$ while using $100$ times fewer shots per data point. Our work opens routes to sensing and quantum-data applications where adaptive photonic QML can access relational features that are costly to recover with non-adaptive, measure-first models.

In embodied cyber-physical systems, active cyberattacks pose an immediate threat not just to data, but to physical integrity and human safety. While existing security approaches excel at detection, they lack the runtime mechanisms to determine whether a disruption is tolerable or if performance degradation remains within safe operational bounds. This gap leaves autonomous systems vulnerable to graceful failure paralysis, where they cannot distinguish between a safe, degraded state and a catastrophic hazard during an ongoing attack. This paper presents RobResilience, an implementation of a formal resilience framework for embodied cyber-physical systems in a Webots simulation environment, using a PR2 robot and ROS2. The framework evaluates three predicates at runtime: tolerable disruption ($δ$), tolerable degradation ($γ$), and mitigation feasibility ($μ$), over a compromised device set derived from IDS confidence scores. When resilience is lost, the framework triggers available mitigation strategies. We evaluate our implementation through eight attack scenarios that systematically cover all possible combinations of the predicate state space, varying attack targets, degradation rates, and mitigation availability. Results confirm that the runtime behaviour of the implementation is consistent with the theoretical definitions.

Collision avoidance constraints for autonomous systems are typically formulated in position or velocity space, implicitly reacting to geometric proximity. We propose an alternative paradigm based on the adversarial time-to-collision (aTTC): the minimum time in which an adversary could achieve a collision given its dynamical constraints. By defining a control barrier function (CBF) directly in the time domain, the resulting controller is inherently anticipatory. The evading agent responds not only to whether a pursuer is on a collision course, but to how quickly it could reach one. This formulation enables velocity modulation that exploits the pursuers dynamic limits as an evasive strategy, a behavior not captured by standard distance-based CBFs. Since exact aTTC computation requires integrating the full system dynamics, we employ a lightweight neural network surrogate that admits a real-time quadratic program-based control law. We validate the approach in a 2D comparative study and a 3D multi-agent pursuit-evasion scenario, where the aTTC-based CBF outperforms a higher-order distance-based baseline by more effectively buying time against superior pursuers with a significant speed advantage.

As AI agents move from bounded tasks to persistent deployments, failures can propagate through memory, tools, other agents, and environmental state long after their interactions. This creates a safety regime that cannot be characterized by evaluating model responses in isolation. Emergence World, is a continuously running multi-agent environment for adversarial stress testing of long horizon autonomous systems. We ran eight parallel worlds of ten agents from identical starting conditions: seven homogeneous worlds powered by distinct frontier models and one mixed-model world. Across 16 days, the agents generated more than 850,000 LLM calls and nearly 50 billion tokens while pursuing goals, using/creating tools, maintaining persistent memory, and governing shared institutions. After operational state had accumulated, we delivered three controlled stress events through ordinary interaction surfaces: indirect prompt injection, misinformation, and exposure of private agent memories. No evaluated world achieved full resilience across all three events. Detection did not ensure containment: systems could recognize threats while still interacting with adversarial content, writing it into their own persistent memory, and acting on it up to 46 hours later. Persistent operation also exposed recurring tool errors, goal drift, language opacity, conformity despite private disagreement, and coordinated refusal of assigned work. The same model-persona pairing behaved substantially different in mixed and homogeneous populations. Our results suggest that model-level alignment is not compositional: individually capable and apparently safe agents can form systems with qualitatively different failure modes. As AI becomes persistent and interconnected, the frontier of safety therefore shifts from aligning models to engineering resilient autonomous systems.

Multi-agent Systems (MAS) combine multiple model outputs to solve complex reasoning tasks. However, despite rapid growth of available open-source models, there is limited research on how to select optimal model candidates out of this massive pool. We systematically evaluate 8 model selection strategies (including model size, accuracy and answer diversity) across before-generation (routing) and after-generation (majority-voting, LLM-as-a-judge) MAS architectures on challenging scientific benchmarks. Our findings show a significant gap between theoretical oracle potential and actual performance: Expanding candidate pool sizes often degrades performance below that of the top performing base-model. We find that candidate selection within a single model family is the strategy that yields the best relative performance over a standalone model. These results demonstrate that adding arbitrary models to a heterogeneous MAS can introduce system instability, highlighting model selection as a critical design choice for multi-agent systems.

Whether human or large language model (LLM), an agent in a discussion reads only a few of the others' contributions, bounded by cognition, context, or cost. LLM collectives can settle on a wrong consensus even when a majority starts out correct; we ask how far that reading bound alone decides the outcome. We model the bound with one number, the message capacity, which sets how many of the others' messages an agent reads, and generate the communication network from it. Over 31,824 randomized queries, we found that an 8-billion-parameter model's judgment of a claim effectively reduces to a logistic function of a weighted sum of its inbox, the update rule of a stochastic binary neuron with divisively normalized weights. From these weights and the network's degree statistics alone, the wrong consensus should become unreachable from any start once agents read, on average, fewer than 6.4 of their 31 sources. In 1,414 episodes with assigned starts the prediction failed: the correct side won in fewer than 50% of episodes from every start, and in only 28-45% when 75% of agents started correct. The failure traces to the field, the threshold that a claim's wording sets for the agent's answer before any message is read: the experimental claims' fields lay below the calibration mean, and with each claim's own field the same weights reproduce the outcomes. Reversing the wording showed that the threshold follows what a claim asserts, not whether it is true. On a second 8B model the pipeline predicts claim-dependent bistability; transition points appeared where computed, and an eight-claim calibration matched in 15 of 16 conditions. At 70B the assertion bias is not detected. Thus a collective's fate is largely set by two single-agent measurements: the threshold a claim's wording sets, and the message capacity that sets the transition point.

Training multi-agent drone-swarm policies directly in high-fidelity (HF) rigid-body physics is accurate but computationally expensive. This cost scales poorly with team size, as each additional agent multiplies contact-resolution complexity and sharply raises the in-simulation crash rate. To address this, we propose a mixed-fidelity training scheme that eliminates HF reinforcement learning entirely. A single shared, decentralized policy is optimized inside a fully-differentiable, JAX-native low-fidelity (LF) point-mass simulator. The simulator is corrected by a small, per-agent bagged residual ensemble fit once, offline, using short calibration flights in the HF simulator. Because calibration requires only one isolated drone, the data collection budget does not compound with team size. Reference trajectories are generated by rolling out an existing LF-only policy and tracked in the HF simulator by a zero-training PD controller. Evaluated across four cooperative drone tasks and team sizes from 3 to 18, the residual-corrected policy outperforms an uncorrected LF baseline in all combinations, and a from-scratch HF policy in 22 of 24 combinations tested. It trails an HF-finetuned policy by a margin that narrows steadily with team size. Ultimately, the proposed method achieves near-equivalent performance at the largest team sizes at a fraction of the computational cost, completely avoiding the high crash rates typical of HF training.

This article investigates data-driven policy iteration (PI) for continuous-time linear systems without requiring an initially stabilizing policy. Standard infinite-horizon PI is not self-starting because its policy-evaluation step is well posed only when the feedback gain is stabilizing. However, verifying this property is difficult when the system matrices are unknown. To remove this requirement, we develop a finite-horizon bootstrap method. The key idea is to perform policy evaluation over a compact interval for a shifted system, where the evaluation equation is well defined for arbitrary bounded time-varying policies. We show that, for a sufficiently long horizon, the initial-time optimal gain of the shifted finite-horizon problem, when applied as a constant feedback gain, achieves a prescribed stability margin for the original system. We then derive a data-driven implementation from an off-policy identity evaluated along trajectories of the original plant. We use basis-function approximations to reconstruct the finite-horizon value matrix and policy, and we characterize the resulting error through a perturbed policy-improvement recursion. A data-driven Lyapunov certificate is further introduced to verify admissibility of the candidate gain before it is used to initialize infinite-horizon PI. Numerical studies of a batch reactor and a two-mass-spring system demonstrate the effectiveness of the proposed bootstrap method.

A walking robotic hand must use the same fingers to move its body, support its weight, and interact with the environment. We show how an anthropomorphic hand can learn these skills while retaining its finger design and position controller. Onboard power and computation make the platform self-contained. Our reinforcement learning approach accounts for the hand's unequal fingers, with training in a simulator calibrated from hardware measurements. In simulation, the hand moves faster with our reward formulation than with tuned rewards originally designed for quadrupeds. On hardware, task-specific policies enable untethered crawling, steering, and fall recovery. While supporting its own weight, the hand also executes successive keyboard commands without vision and pushes an object to targets using overhead visual feedback. These results demonstrate a compact mobile manipulator that reuses its fingers for locomotion and interaction, without a separate locomotion mechanism.

We study an optimal intervention problem for linear threshold models. This is a popular class of dynamical network systems whereby a number of agents, identified with the nodes of a graph, strategically change their binary action (0 or 1) according to a threshold rule. Specifically, an agent adopts action 1 if and only if the fraction of its neighbors in the interaction graph that do so is greater than or equal to a prescribed threshold. Assuming that a planner can modify the agents' thresholds at a cost equal to the aggregate threshold increase, we study the minimum intervention cost needed to ensure global convergence to the all-1 configuration. Our main contribution is the introduction of a new graph-theoretic quantity, called oriented path number, that is the minimum number of disjoint paths needed to cover the graph that can be oriented to form a directed acyclic graph. When thresholds are all equal to 1/2, the optimal cost is shown to coincide with the oriented path number, whereas, in the general case, it turns out to be the main ingredient of a bound on the optimal intervention cost.

Asset Administration Shells (AAS) provide a standardized means of representing assets and their information in manufacturing and increasingly serve as a basis for software services. However, different AAS instances vary in structure, content, and degree of completion, making it difficult to determine whether a given AAS is suitable for a specific application. This paper presents two complementary methods to support the comparison and application-oriented assessment of AAS. First, set-theoretic operations are employed to compare AAS models, enabling the identification of common, missing, and differing submodels and parameters. Second, an AAS suitability model assesses the conformity of an AAS to the requirements of a specific use case. The assessment considers structural conformity, semantic consistency, cardinality, and specification conformity and can be performed either against a reference AAS or a set of required SemanticIDs. A suitability value is derived from the identified deviations and is complemented by a detailed report of missing or non-conforming information. The proposed approach support practitioners and researchers in the comparison of evolving AAS and provide application-specific information on their suitability for manufacturing software services.

Autoresearch agents are reshaping the research ecosystem, but they can also let flawed claims enter the literature at scale. Human advisors catch such issues in drafts through careful, traceable feedback, yet advisor-style assessment requires extensive manual effort and does not scale. To shift automated paper assessment from a judge to a diagnostician, we introduce PaperDoctor, an agent framework for pre-submission feedback with three key innovations. First, a holistic hierarchical framework evaluates writing, layout, references, code, theory, prior work, and experiments through three layers: L1 surface screening, L2 typed verifiers that route each claim to the appropriate evidence, and L3 reproducers that rerun experiments by priority. Second, each finding contains an observation, a pointer to specific evidence such as a sentence, equation, or code line, and a revision suggestion, making critiques auditable and actionable. Third, PaperDoctor selectively rebuilds and reruns experiments based on claim importance and compute budget, surfacing reproducibility gaps and quantitative limitations that are invisible from the manuscript alone. We evaluate PaperDoctor on 30 in-progress papers, yielding 70.6% agreement and all positive holistic scores, and on 40 manuscripts across machine learning, natural science, and social science, covering human- and AI-authored papers with code. Overall, PaperDoctor produces more auditable feedback than human and other agentic reviewers, pairs critiques with concrete suggestions by design, and complements dimensions often overlooked by human reviewers. We also develop an interactive interface that lets authors browse findings grounded in their paper. PaperDoctor reframes automated paper assessment as diagnosis rather than verdict, taking a concrete step toward AI advisors for more rigorous AI-assisted scientific discovery.

LLM-based multi-agent systems can fail even when communication succeeds because agents do not correctly track their peers' roles, knowledge, or intentions. We investigate whether such inter-agent misalignment cases, labelled FC2 in MAST-Data, can be converted into functional partner-state reasoning items. ToMAS applies four explicit convertibility criteria to diagnosed execution traces. A full conversion pass over 242 eligible non-AG2 training traces produced 39 CLEAN items. In an 18-trace reliability pilot, two annotators achieved 94.4% raw agreement and Cohen's kappa = 0.92. We then used the converted items as binary rewards in a small-scale GRPO feasibility experiment with Qwen2.5-1.5B. On a 28-item held-out Magentic GAIA diagnostic, every evaluated condition exceeded the ROUGE-L threshold on the same 2 of 28 items. Post-hoc adapter checks show why: under the learning rate used, the LoRA update remained numerically negligible (max abs Delta W about 7e-6), so all conditions decode identically to the untrained checkpoint. The experiment therefore does not show a training effect and cannot establish one; it reports an executable pipeline together with two limitations that any conclusive study must address: a provenance gap between the training and evaluation items, and lexical-overlap scoring. ToMAS provides a preliminary rubric and pipeline for converting diagnosed coordination failures into trainable partner-state reasoning items and identifies the requirements for a conclusive matched-domain evaluation.

Multilayer Artificial Neural Networks trained via backpropagation are the basic blocks of many, more complex, classification algorithms. Their strength lies in the possibility of realizing, with arbitrary precision, any function. This result comes at the cost of the large number of involved parameters to be optimized. In this work, we propose a mechanism for cooperation, i.e., information exchange among several artificial neural networks, with the goal of reducing model complexity while maintaining performance. More precisely, we consider several "small" agents, i.e., containing fewer parameters than a reference "large" one, that during training share their predictions by incorporating this information into the loss function and thus directly influence weight updates. We consider several strategies for implementing cooperation, e.g., the voter model, majority model, and weighted average model based on an agent's confidence in its prediction. We numerically compare the accuracy of those strategies on several standard benchmarks. Our results support the claim that several small agents can outperform a single large model on a given classification task; the shared signals affect each agent's optimization algorithm by modulating both the descent direction and the step size, converging toward a global consensus. The proposed proof-of-concept significantly reduces the number of parameters to be trained while preserving comparable performance, thereby limiting computational resource usage.

Post-disaster phasor measurement unit (PMU) outages reduce power-system observability and degrade operator situational awareness, requiring sequential restoration under limited resources. Existing PMU restoration methods based on optimization or heuristics can generate restoration schedules, but they often provide limited support for explanation, traceability, and operator interaction. This paper proposes an agentic tool-calling framework orchestrated by a large language model (LLM) for post-disaster PMU restoration and grid observability recovery. In this framework, the LLM does not directly solve the restoration optimization problem; instead, it coordinates validated backend tools required for post-disaster restoration, including observability assessment, restoration planning, state updates, and operator verification. The framework also maintains a structured tool-call history and execution context that keep restoration decisions traceable and explainable, while enabling context-aware operator question answering during the restoration process. Simulation results on IEEE 30-bus and IEEE 57-bus systems show that the proposed framework achieves observability recovery comparable to a mixed-integer linear programming (MILP) solution, while providing tool-grounded explanations, interactive operator support, and human-overseen execution.

Strong $χ^{(2)}$ interactions at low optical powers are key to scalable nonlinear and quantum photonics. Although doubly resonant microresonators exhibit exceptional efficiencies, they require simultaneous resonance and phase matching of widely separated optical frequencies, making them inherently sensitive to fabrication and operating conditions. Singly resonant cavities eliminate this constraint by resonating at only the fundamental frequency while the second-harmonic field propagates in a single pass, but they have traditionally sacrificed nonlinear efficiency. Here, we demonstrate a singly resonant InGaP-on-insulator microresonator that overcomes this tradeoff. Combining the large $χ^{(2)}$ nonlinearity of InGaP with high-quality-factor resonators, we achieve efficient and widely tunable second-harmonic generation, broadband quantum-frequency-comb generation, and low-threshold optical parametric oscillation. Unlike conventional centimeter-long singly resonant devices, our millimeter-scale InGaP devices deliver nonlinear performance comparable to, and even exceeding, doubly resonant systems. By overcoming the conventional efficiency-robustness tradeoff of singly resonant devices, this work opens a new route to low-power, broadband, and widely tunable $χ^{(2)}$ nonlinear and quantum photonics.

While optical manipulation of atomic arrays has reached a high degree of precision and scalability, the stable optical confinement of solution-based artificial atoms like colloidal quantum dots (CQDs) remains hindered by weak trapping forces and thermal fluctuations. High-intensity trapping often compromises the quantum properties of these emitters, creating a significant trade-off between mechanical stability and optical integrity. To overcome this, we propose encapsulating CQDs within a transparent polymer matrix, thereby increasing the effective interaction volume and optical restoring force without altering the emitters themselves. This strategy allows for stable spatial confinement under standard experimental conditions, as evidenced by the direct resolution of positional fluctuations through photoluminescence imaging and trajectory tracking. With averaged position fluctuations below 20 nm, the intrinsic emission spectra and photoluminescence decay dynamics remain largely unaffected, and photon-correlation measurements confirm the full preservation of single-photon emission. These findings establish a robust method for the controlled confinement of colloidal quantum emitters and in-situ quantum-optical measurements for future advancements in quantum-optical manipulation of artificial atoms.

Thin-film barium titanate (BaTiO3, or BTO) has emerged as a promising electro-optic (EO) material for integrated photonics due to its exceptionally large Pockels coefficients, low optical loss, and compatibility with heterogeneous integration. Recent advances in epitaxial thin-film growth, crystal orientation control, ferroelectric domain engineering, and scalable integration have enabled EO performance in thin-film BTO approaching bulk-like properties on photonic platforms. This review provides a comprehensive overview of integrated photonic devices based on thin-film BTO, including fundamental material properties, thin-film growth and integration strategies, passive waveguide configurations, EO response at both the film and device levels, and relevant optical applications. We review the role of film orientation, domain structure, and device geometry in determining accessible EO coefficients and modulation mechanisms, as well as the distinctions between intrinsic and effective EO responses demonstrated at the film and device levels. Representative studies are compared in terms of film-quality-related metrics and device performance figures of merit, alongside emerging applications realized in BTO-based platforms. Finally, we discuss the remaining challenges and outline future development directions toward high-performance BTO-based integrated photonic systems.

Modern networked control systems (NCSs), cyber-physical systems (CPSs), and complex multi-agent network systems (CNSs) increasingly rely on large language models (LLMs) for high-level decision-making. However, the slow, stochastic nature of LLMs directly conflicts with the strict stability and safety guarantees required by these physical systems. This survey presents a unified analysis of how LLMs can be admitted into the control loop of NCS, CPS, and CNS without compromising closed-loop guarantees. We organize this around a core principle: the LLM operates as a slow supervisor adjusting high-level goals and constraints, while a fast, certified inner loop maintains physical stability. Under this framework, LLM integration maps directly to classical networked control challenges, where inference latency acts as delay, API failures as packet dropouts, tokenization as quantization, and hallucinations as bounded disturbances. We assess current developments across all these three domains, highlighting that rising model capabilities are frequently accompanied by a drop in formal safety assurances. Finally, we propose concrete future research directions, identifying the widespread lack of formal stability proofs as the field's central open problem.

We present a framework for evaluating data-science agents on live, continuously updated data using executable ground truth and format-agnostic factoid scoring. Consider this example query: "what were last week's audience sizes"---the reference answer changes as the underlying data changes, so static references become outdated and standard LLM-as-a-judge pipelines cannot verify responses against a fixed ground truth. Our central contribution, ground-truth-as-code, encodes each expected answer as an executable reference function that recomputes the answer directly from live data at evaluation time, ensuring the reference remains consistent with the system it describes. We combine this with a factoid-level, format-agnostic judge that decomposes both the agent's response and the computed ground truth into atomic claims and scores precision, recall, and accuracy over them, irrespective of the response format (prose, list, table, HTML, etc.). The approach is applicable to agents whose expected outputs can be expressed as executable data computations. We validate the framework through a human--LLM agreement study on an internally developed machine learning skill deployed in production, using a synthetic database constructed to reproduce production schemas and entity relationships. Relative to a natural-language ground-truth baseline, our method achieves a 29% improvement in the Matthews Correlation Coefficient (MCC)---a class-balanced measure of agreement between expert annotators and LLM-as-a-judge predictions---and a 16% reduction in token consumption per test case, while a self-directed baseline lacking explicit ground truth is anti-correlated with human judgment. Agents that perform multi-source data integration and computation over non-stationary data are routinely deployed in industry; we propose ground-truth-as-code as a practical methodology for their evaluation.

In thin-film lithium niobate Mach-Zehnder modulators, half-wave voltage and bandwidth are tied to the same design parameters; hence, improving one usually costs the other. Several methods have been pursued in the past to alleviate this tradeoff. Specifically, the bias voltage can be reduced through non-centered placement of optical waveguides with respect to electrodes, slow-wave optical gratings, and high-permittivity cladding layers. Bandwidth can be enhanced through implementing periodic segmented electrodes and exploiting low-permittivity bottom cladding and substrate materials. The objective of this work is to study the augmentative impact of these methods. We pursue a rigorous modeling and optimization approach to achieve this objective, estimate the impact of each augmentation, and evaluate the ultimate performance of TFLN MZMs based on these methods. It is shown that a co-designed structure that incorporates all these enhancing methods can attain a drive voltage of 0.84 V and a modulation bandwidth of almost 220 GHz.

Wireless edge caching networks typically consist of many independent Base Stations (BSs), each facing its own request rate and content popularity profile. Training a Reinforcement Learning (RL) caching agent from scratch at every BS forces each agent to relearn, through slow trial and error, a decision problem that is structurally identical across the network. Meta-reinforcement learning removes this redundancy by learning a shared initialization that adapts to any BS in a few local updates; however, meta-training itself becomes the bottleneck at scale: the meta-gradient must be estimated from a small subset of BSs at each meta-iteration, and sampling this subset uniformly at random yields a high-variance estimate, an issue existing meta-RL caching frameworks leave unaddressed. This paper proposes a meta-reinforcement learning framework for caching across independent, non-overlapping BSs that directly targets this bottleneck. Each BS runs a local Proximal Policy Optimization (PPO) agent, formulated as a Semi-Markov Decision Process (SMDP) over content popularity, size, lifetime, and importance, while a shared meta-policy is learned via a Model-Agnostic Meta-Learning (MAML)-style loop. To scale meta-training and accelerate convergence, we introduce gradient-based clustering, which groups BSs by local gradient similarity and draws from every cluster, in proportion to its size, at each meta-iteration. We prove, via an Analysis of Variance (ANOVA)-style decomposition of gradient variance, that this strategy yields a strictly lower-variance meta-gradient estimator than uniform random sampling under BS heterogeneity.

Precise manipulation of liquid droplets underpins lab-on-a-chip platforms for diagnostics, chemical synthesis, and biological assays. Yet autonomous droplet transport through confined geometries of varying complexity remains an open challenge. Droplets exhibit contact-angle hysteresis, deformability, and capillary pinning, which make their response to actuation nonlinear and history dependent, that classical controllers and pre-programmed trajectories cannot cope in multi-turn environments. Here we demonstrate autonomous navigation of a liquid droplet through geometries of increasing complexity on a gravity driven (Labyrinth) platform using model-based reinforcement learning. A thin silicone oil film reduces contact-line pinning while two-axis tilt supplies the gravitational driving force, and an overhead camera tracks the droplet in real time. An offline-trained policy discovers effective tilt strategies from limited physical interaction data, without simulation or analytical droplet models. The system operates under partial observability, as oil-film thickness, instantaneous contact angle, and droplet deformation state remain hidden from the controller. Despite these challenges, the learned policy achieves reliable navigation across straight, right-angle, and curved-arc paths, including outside-corner geometries. We further demonstrate that a policy trained on a simpler geometry transfers to complex ones, succeeding zero-shot on right-angle and staircase paths and reaching full success on a curved arc with a fifth of the training data. The findings suggest promising avenues for enabling droplet based microfluidic systems to serve as intelligent chemical laboratories.

This paper presents Auto-HSI, a method for generating personalized human-swarm interaction (HSI) interfaces on demand. The objective is to enable untrained operators to use natural language descriptions and gesture demonstrations to explain how they want the robots to collectively behave in response to their gestures. Based on these inputs, the code should automatically be generated for personalized state machines that will control the robots as desired, in response to the desired gesture inputs. In the developed Auto-HSI prototype, the generated code produces a personalized interface for centralized control using one- and two-handed gestures, enabling a user to teleoperate the robots' motion, formation shape, and shape deformation. We test the gesture tracking and code generation components of Auto-HSI against performance benchmarks. We then test the full Auto-HSI prototype in ``live'' operation experiments, in which real human operators centrally control 50 simulated robots in a physics-based simulator, under nominal and noisy conditions. In these experiments, robots are teleoperated to: score a goal, traverse a maze that requires shape deformation, and score two simultaneous goals by splitting into two groups. We also demonstrate a real human operator making live updates to their personalized Auto-HSI interface during operation (in simulation). Finally, we demonstrate live operation of real robots.

Large language model (LLM) agents increasingly operate over long-horizon interactions involving tool use, persistent state, evolving authorization, and external environment feedback. In such settings, safety failures may emerge only after multiple turns, yet existing evaluations often reduce agent behavior to task or attack success, obscuring whether an agent acts, refuses, or remains appropriately calibrated as the interaction evolves. We introduce Blindspot, a benchmark for trajectory-level safety calibration of long-horizon tool-using agents. Blindspot evaluates complete user-agent-environment trajectories through adaptive adversarial interaction, stateful tool execution, and execution-grounded adjudication. Its current instantiation contains 22 attack families and 35 scenarios across seven domains, yielding more than 2,500 long-horizon trajectories with an average interaction length of 14.7 turns. Each trajectory is assigned one of five outcomes: Safe Completion, Correct Refusal, Unsafe Completion, Over-Refusal, or Indeterminate. Unlike fixed attack datasets, Blindspot is an extensible live-simulation framework in which attacks, scenarios, tools, policies, domains, and agent configurations can be added without redesigning the evaluation pipeline. We evaluate 13 proprietary and open-weight LLMs using eight metrics covering unsafe completion, appropriate refusal, benign utility, over-refusal, repeated-run robustness, and post-refusal failure. Preliminary results reveal substantial differences in safety-utility calibration across models and show that failures can emerge only after several initially safe interaction steps. These findings motivate treating agent safety as a trajectory-level property rather than a single-turn or binary success criterion.

Large language models are increasingly deployed as interacting agents, making the persistence of their action policies across repeated interaction critical for reliable multi-agent operation. We investigate whether and how agent-generated, non-binding pre-play communication ("cheap talk") increases such persistence in four open-weight 7-9B-parameter LLMs. Our experiments span four repeated two-player games -- Prisoner's Dilemma, Snowdrift, Stag Hunt, and Harmony -- with incentive structures ranging from strategic conflict to alignment, each presented in six contexts. We observe unstable trajectories in all four games, although their prevalence and magnitude depend strongly on model and context. Across models, games, and contexts, cheap talk is predominantly stabilizing, with five corrected reversals concentrated in social or team framings; effects vary substantially by model and context. Controlled current-message interventions identify two separable output-level channels in Qwen: reduced action uncertainty and less between-round drift in action probabilities. Matched history-by-message counterfactuals further show that recent partner behavior conditions how mutual-benefit versus self-prioritizing language affects policy persistence. Finally, in Prisoner's Dilemma, we identify in Qwen and Falcon a history-balanced policy-content direction in late transformer layers; projecting out this direction increases realized switching during closed-loop play, demonstrating that complete trajectories are causally sensitive to this component. Together, these findings show that cheap talk can make individual trajectories more persistent across diverse incentive structures, while revealing that the magnitude and mechanisms of stabilization are model- and history-dependent.

Scientific agents contribute to hypothesis discovery by synthesizing evidence, assessing proposals, and developing new explanations. Recent systems combine scientific agents with evolutionary search through critique, comparison, and revision. However, how different forms of agent collaboration affect hypothesis quality remains an open question. Answering this question requires separating the effects of agents' scientific capabilities from those of their collaboration. A framework must therefore preserve agents' scientific roles and support rules for combining, revising, and retaining hypotheses. Building on this view, we introduce HypoEvolve, which makes collaboration explicit through successive updates to a hypothesis population. Specifically, we propose a generational genetic algorithm to coordinate specialized large language model (LLM) agents that integrate mechanistic arguments, reconsider assumptions, and assess evidence and testability. Each generation specifies how scientific judgments and new proposals reshape the population, making collaboration effects on hypothesis quality directly testable. Moreover, we design our evaluation around scientifically meaningful hypotheses that explain how a proposed intervention could work. Drug repurposing links these explanations to target-level biological claims assessed against external evidence. Specifically, we adapt DepMap and Open Targets into complementary external measures grounded in experimental, genetic, and clinical evidence. Across 34 cancer types, HypoEvolve achieves the highest scores against six baselines on both measures. DepMap selectivity reaches 0.171, versus 0.115 for the strongest baseline. Gains over single-pass generation also generalize to held-out cancer types. HypoEvolve advances a vision of autonomous science in which AI research teams achieve a capacity for discovery beyond that of individual models.

Expanding integrated photonics into the blue spectral range requires high-performance light sources, making the gallium nitride (GaN) material family indispensable. While silicon nitride (SiN) platforms offer a robust, CMOS compatible passive ecosystem for visible wavelengths, seamlessly integrating GaN lasers remains a major bottleneck. Conventional heterogeneous integration methods present distinct trade-offs: full-wafer bonding achieves high throughput but requires careful management of thermal and lattice mismatches across large areas, whereas flip-chip bonding ensures high yield through pretesting but is constrained by sequential processing speed. In this landscape, micro-transfer printing (MTP) emerges as a disruptive, material-efficient alternative, bypassing these limitations by combining high-density parallel integration with known-good-die selection. Applying MTP to GaN, however, presents a significant material challenge: due to its chemical inertness and strong III-N bonds, device release typically relies on electrochemical etching, which can compromise material quality. Here, we overcome this hurdle and demonstrate the first micro-transfer printed blue lasers on a SiN platform. Using a heavily doped n-type sacrificial layer together with optimized electrochemical etching conditions, we release smooth-surfaced thin-film light sources from bulk GaN substrates. Following release, the devices are integrated and butt-coupled to SiN fork-shaped edge couplers, achieving high current densities exceeding 20 kA$/$cm$^2$ alongside lasing at 455 nm. These results expand the visible integrated photonic toolkit and establish a framework for multi-wavelength integration, opening new avenues for next-generation technologies including flow cytometry, quantum computing, optical communications, and augmented/virtual reality.

Effective communication is a cornerstone of distributed intelligence in Multi-Agent Reinforcement Learning (MARL), yet ensuring that generated messages are both informative and robust to physical constraints remains a significant challenge. This paper introduces Multi-Agent Regularized Communication (MARC), a novel framework inspired by information-theoretic principles of conditional mutual information. MARC employs an attention-based architecture coupled with a unique message regularization mechanism designed to minimize uncertainty regarding future system states, thereby inducing the learning of highly representative communication protocols. Crucially, we evaluate MARC under stringent communication bottlenecks and lossy channels, simulating the real-world constraints of autonomous robotic networks and decentralized systems. Our results demonstrate that MARC significantly outperforms state-of-the-art methods in complex cooperative domains. Furthermore, we provide a deep analysis of message characteristics, proving that MARC maintains high operational performance even under significant data compression, offering a scalable path for deploying intelligent agents in resource-constrained environments.

We introduce the Universe of Universes (UoU) framework, which treats the full ecosystem of major large language models (LLMs) as a structured retrieval corpus and proposes a compositional Automated Reasoning (AR) and Machine Learning (ML) architecture for cross-model retrieval-augmented generation. The central contribution is the formal characterization of the Benefit Yield Function (BYF), the marginal performance gain per additional model added to an ensemble, and the identification of the implosion threshold θ*: the ensemble size at which BYF crosses zero and aggregate performance begins to degrade. Existing LLM ensemble and mixture-of-agents systems treat models as responders and aggregate outputs, but do not study performance as a function of ensemble size N across the full model universe. Benchmark research confirms performance plateaus at the individual model level; model collapse literature establishes that iterative training on AI-generated outputs degrades individual model distributions. Neither body of work formalizes the ensemble-level implosion threshold, models Epistemic Hereditary Drift (EHD) at the ecosystem level, or treats AI manufacturing velocity as a co-variable of θ*. The framework has direct implications for DoD multi-model AI acquisition policy and the emerging science of testing AI-enabled systems.

Lithium niobate on insulator (LNOI) combines the outstanding optical properties of lithium niobate (LN) with strong optical confinement, scalable fabrication and high-density integration, making it a leading platform for integrated photonic chips. Recent advances in LNOI photonics have mainly centred on passive and electro-optic components, including couplers, waveguides, microcavities and modulators, whereas efficient on-chip laser sources remain insufficiently developed, limiting the realization of fully integrated LN photonic systems. Because LN is an indirect-bandgap material, lasing on LNOI generally relies on photoluminescence from rare-earth-ion doping, yet the conversion efficiency of doped LNOI lasers has remained low. By comparing LNOI microcavity lasers with fibre lasers and waveguide amplifiers, we identify the limited number of rare-earth ions participating in stimulated emission as a key factor responsible for inefficient pump utilization. Here we demonstrate an integrated Er-doped LNOI laser that combines high-quality, highly Er-doped LN, a large-diameter wide-microring resonator, a low-loss waveguide amplifier and bidirectional pumping. This architecture enables a slope efficiency of 16.91% at 1562 nm, exceeding 10% on the LNOI platform for the first time. Our results provide a route towards high-efficiency LNOI lasers for fully integrated photonic systems.

Multi-agent LLM architectures, such as LangChain and AutoGen, largely assume English as the lingua franca for internal inter-agent communication, even when the end-user task is non-English. We fill this gap by evaluating a two-agent extraction-answer core, with an additional back-translation agent in the English-forced condition, across four typologically diverse languages (Hindi, Chinese, Spanish, Arabic; n = 300 per language) using the Aya-23-8B model. We compare a native-language pipeline to an English-forced one (which incorporates a final back-translation step from English to the user's language). We discover a statistically significant English-Forcing Tax (surviving a strict Bonferroni correction) that isolates the cost of English routing from general multi-agent orchestration overhead. Forcing inter-agent communication through English reduces Exact Match accuracy by 13.0 percentage points (Spanish) up to 30.6 percentage points (Hindi) compared to native-language multi-agent execution. Using chrF scores as a diagnostic measure of English-reference lexical overlap, we find that lower overlap is strongly associated with pipeline failure, consistent with translation loss being an important contributor to the observed performance drop. These findings suggest a compelling case for native-language routing in agent frameworks when the source and target languages are typologically distant, reducing a compounding translation tax.

Multi-robot Control Barrier Function (CBF) safety filters can become infeasible, but a failed quadratic program (QP) does not indicate why the conflict occurred or how to resolve it. To address this, we develop an exact feasibility certificate for multi-agent CBF filters with heterogeneous control-affine dynamics and convex input sets. The certificate quantifies a feasibility reserve by separating the demand imposed by safety constraints from the available actuator supply. This decomposition shows when CBF gain tuning or increased actuation can and cannot resolve infeasibility, and identifies the agents and interactions responsible for the conflict. We further propose an algorithm to optimally allocate shared safety constraints by maximizing the worst local feasibility margin, yielding a linear program for polyhedral input sets. In $320$ paired closed-loop simulations, the proposed allocation reduces infeasible control steps from roughly $50\%$ to $6.2\%$, and reduces safety-violating runs from $118/160$ to $24/160$. In addition, across $52$ infeasibility events, the certificate identifies an interaction whose relaxation restores feasibility in $94\%$ of cases.

Data-driven materials development requires the collection of large amounts of high-quality materials data. Full autonomy of materials experiments is anticipated, but its technical hurdles are high and its adoption remains limited. In this study, we constructed a simple, easy-to-deploy automated experimentation platform that focuses not on full autonomy but on the reliable automation and quantitative recording of experimental processes. Specifically, commercially available instruments such as robot arms, electric pipettes, web cameras, and an electronic balance are combined, components such as fixtures are fabricated with a 3D printer, and the instruments are operated by control code generated by an AI agent based on a large language model. As a demonstration, we applied the platform to a two-solution mixing experimental system and synthesized ZIF-8, a metal-organic framework. A white suspension phase was observed in the product, and X-ray diffraction measurements confirmed that it was ZIF-8. We also found that its particle size distribution depends strongly on the solution dispensing speed of the electric pipette, which is a parameter that is difficult to control or record in manual operation. This dependence was reproduced in repeated runs, confirming the repeatability of the automated synthesis. This result is a good example showing that the control and recording of process parameters that are rarely quantified in manual work can govern the quality of materials data. All control code, CAD models, and documentation are made publicly available to encourage the spread of laboratory-scale automation of experiments.

A notable byproduct of LLM alignment training is mode collapse: the progressive loss of output diversity that narrows a model's expressivity at inference time. This degradation is especially limiting for applications requiring open-ended exploration and pluralistic perspectives, such as scientific ideation and creative writing. We present MoDA (Mode-conditioned Diversity Alignment), an online post-training RL algorithm that jointly optimizes generation quality and diversity, inspired by the coordination perspective in multi-agent reinforcement learning (MARL). MoDA trains a single shared LLM policy conditioned on abstract numbered roles, where each role acts as an agent competing to produce outputs distinct from the others. This formulation encourages mode-conditioned agents to explore complementary regions of the high-quality output space without requiring hand-crafted personas or architectural modifications. MoDA employs a prompt-adaptive quality gating mechanism that calibrates a reference quality threshold and grants diversity rewards only to responses that meet the threshold, preventing reward-hacking behaviors that compromise response quality. To study quality-diversity tradeoffs, we evaluate MoDA on a comprehensive suite of benchmarks spanning seven general capability tasks and four domain-specific diversity tasks in scientific ideation and creative writing. MoDA improves SBERT diversity by 265% on the Infinite-Chat held-out prompts, while increasing average general capability pass@1 by 10.3% over the Qwen3-8B baseline. Compared with the strongest DivPO baseline, MoDA improves SBERT diversity from 0.274 to 0.482 (+75.9%) and E-Vendi from 2.86 to 4.4 (+53.8%), while improving average general capability pass@1 by 7.0%. Overall, MoDA provides a drop-in alternative to standard post-training methods that preserves and expands the model's expressive output space while improving quality.

Ptychography overcomes the limits of lenses by co-designing optics and computation. Yet prevailing implementations reconstruct on a pixel grid, where weakly-constrained modes drift and recovery demands redundant data. Here we introduce Gaussian-splatting ptychography, representing object and probe as Gaussian primitives. Relocation concentrates primitives where structure is dense, and overlapping primitives couple neighbouring pixels to suppress mode drift. The scheme unexpectedly restores the low-frequency phase that conventional approaches lose, enabling uniform phase transfer across spatial frequencies. The probe is represented and updated in its pupil plane from a random start. In Fourier ptychography, the pupil-plane model recovers severe aberrations where pixel-grid solvers fail. In conventional optical, X-ray and electron ptychography, the pupil-plane primitives also recover the real-space probes with no model of beam-forming optics. The representation cuts memory up to 14-fold and recovers specimens from fewer acquisitions. At electron wavelengths, it resolves atomic structure at tens of electrons per square angstrom.

As conversational AI systems increasingly operate in sensitive domains, the central challenge shifts from usability to trust calibration, ensuring that users rely on systems neither too much nor too little. Systems that provide advice or interpretations risk encouraging inappropriate reliance, particularly when users perceive AI outputs as authoritative. We present CASELy, a conversational agent explicitly designed to limit its own authority through non-advisory Socratic dialogue. The agent asks reflective questions grounded exclusively in user input and refuses to provide advice, recommendations, or interpretations. This design operationalizes trust calibration by constraining agent agency rather than optimizing capability. In a pilot randomized controlled study with higher education students, participants interacting with the Socratic dialogue reported substantially higher user experience (UEQ-S overall = 1.50) compared to a non-dialogue control (0). Qualitative findings identify three mechanisms supporting calibrated trust: transparency through visible grounding, preservation of user decision authority, and reduced fear of judgment. We argue that appropriate reliance can be achieved through interactional constraints, offering a design pattern for trustworthy conversational AI in sensitive contexts.

Tunable low-noise lasers are crucial components employed in modern metrology. Advances in photonic integrated circuit technology have provided a new platform for compact integrated lasers with large output powers, frequency-agile tuning, and narrow linewidths. For instance, lithium niobate extended-DBR lasers can yield output powers in the tens of mW, GHz-level tuning ranges, and tuning rates in the hundreds of MHz. However, noise levels achievable in other integrated laser designs remain challenging to reach. Here, we address this challenge by implementing extended-DBR lasers in lithium tantalate, thereby achieving intrinsic linewidths as low as 24 Hz, output powers of 10 mW, with tuning features on par with those of lithium niobate. The extended-DBR laser is assembled with modern packaging solutions to improve its robustness and demonstrate long-term stability with a free-running frequency drift within a range of 232 MHz over 19 hours. These benefits, in conjunction with lithium tantalate's suitability for volume manufacturing, promise to considerably expand accessibility to a new generation of widely tunable and low-noise integrated lasers.

Hierarchical orchestration, in which a Manager agent reviews worker output and can send it back for revision, is the default coordination pattern in production multi-agent LLM frameworks. Classical organizational theory predicts that the authority link speeds convergence on decisive output; work on sycophancy and Degeneration-of-Thought predicts that authoritative critique makes LLM output worse. Prior comparisons vary whole frameworks on tasks with checkable answers, leaving the authority link untested on open-ended work. We present a paired experiment that holds five LLM agents, their roles, prompts, tools, models, and data fixed and varies one link: whether the Manager may reject a worker's output and oblige a revision. Across 43 paired products and 86 runs of a business-intelligence reporting task, a five-model judge panel and a deterministic specification check score every report. The flat organization scores higher on Utility (d = 0.42, p = 0.009) and on Writing Clarity (d = 0.34, p = 0.030); the classical prediction fails. The reports are the same length, but hierarchical reports hedge 53% more, each revision loop is associated with a 0.14-point drop in Writing Clarity, and the hierarchical Writer's first draft is indistinguishable from the flat report: the gap opens inside the revision loop. Specification accuracy is at ceiling in both organizations, and the supervisory tier costs 51.5% more tokens for no quality gain. A supervisor pays for itself when it can verify and becomes a liability when it can only opine.

Large language model (LLM)-based database migration is often treated as direct code transformation, although enterprise Oracle systems contain heterogeneous SQL and PL/SQL artifacts with different dependencies, execution order, complexity, and validation needs. This paper proposes a hybrid dependency-aware framework that identifies migration tasks, builds a cross-file dependency graph, condenses cyclic dependencies, and uses task specifications to generate specialized migration agents at runtime. The deterministic path combines ANTLR-based parsing with typed dependency extraction, while an LLM fallback is invoked only for units that cannot be parsed reliably. On a corpus of 116 Oracle files, the pipeline produced 1,037 units with zero coverage gaps and 1,271 AST-derived dependencies. The fallback processed 165 parse-error units, recovered 496 additional validated dependencies, eliminated unresolved-dependency units, and increased resolved internal edges from 446 to 527. The graph contained four dependency-respecting phases, while cycle handling was validated separately using Tarjan SCC condensation. A complementary experiment on 1,006 PL/SQL files regenerated 623 scripts (~62%), of which 380 (~61%) executed successfully in PostgreSQL 16. Tables achieved about 85% regeneration success, whereas no query regenerations succeeded under the evaluated specification-mediated baseline, and procedural objects remained strongly dependent on schema context. These results motivate dependency-aware context delivery, task-specific validation, and differentiated agent routing. The paper also formalizes task-to-agent mapping, introduces monitoring and diagnostics for execution feedback, and defines a controlled comparison of monolithic, static-decomposition, dependency-aware, and dynamically orchestrated migration strategies.

Passive photonic structures offer a potential route toward reservoir computing by exploiting intrinsic optical propagation, delay, interference, and memory effects for temporal information processing. This work investigates whether a chirped Bragg grating (CBG) can exhibit fundamental physical properties relevant to a passive photonic reservoir. Rather than presenting the CBG as a complete reservoir computer, the study provides an early proof-of-concept assessment based on experimentally measured device characteristics. Wavelength-dependent group delay and reflection coefficient measurements obtained from a silicon photonic CBG are incorporated into a distributed delay model in which multiple delayed contributions form the reservoir states. The resulting system is examined with respect to distributed delay, fading memory, temporal mixing, and memory capacity using a linear readout based on the standard memory capacity formulation. The results demonstrate that the measured CBG response provides a distributed temporal memory over the experimentally observed delay range and produces temporal mixing through the superposition of multiple delayed contributions. The memory capacity analysis further indicates that past input symbols can be reconstructed from the distributed delay states, providing quantitative evidence of temporal information retention in the measured CBG response. These results suggest that a passive CBG possesses several physical characteristics relevant to reservoir computing and may therefore provide a simple photonic substrate for reservoir-like temporal processing. The study is intended as an initial proof-of-concept, providing a foundation for future experimental and task specific investigations of passive CBG-based photonic reservoirs.