Physical AI Brief

Existing scaling strategies for Multimodal Large Language Models (MLLMs) typically expand either model parameters or sequential inference computation, incurring substantial memory or latency overhead. More importantly, most existing methods fail to alter the rigid, fixed computation allocation between the Vision Transformer and the Large Language Model components, limiting task-specific optimization. To address this, we introduce the Parallel Vision-Language (ParVL) scaling framework for MLLMs, which scales parallel computation by reusing the existing ViT and LLM backbone parameters across multiple vision and language branches. This framework raises a central question: given a fixed backbone parameter budget, how should additional shared-backbone computation be allocated between the vision and language modalities? We instantiate each parallel computational stream with branch-specific prefix parameters over a shared backbone, and train the entire model end-to-end via full-parameter supervised fine-tuning on roughly 13B tokens. We systematically study the computation-allocation trade-off between the ViT encoder and LLM decoder. ParVL improves overall multimodal performance over same-recipe single-branch baselines, and the best evaluated vision--language allocation varies across tasks. Code is available at https://github.com/YangYangGirl/ParVL.

Large language models (LLMs), and the agents built on top of them, are now benchmarked heavily on whether they can finish a task -- fix a bug, drive a browser, operate a GUI. A complementary social ability, namely how well a model understands and forecasts the way real social events unfold, has barely been measured. We introduce SocietyBench, an end-to-end benchmark that takes a one-line event topic, collects Web news and social-media posts across five platforms, distills them into a date-indexed timeline that keeps factual events and a public-opinion layer separate, and then turns every cutoff date on that timeline into an audited bank of forecasting questions. Questions are scored on two orthogonal 100-point axes: probability calibration and temporal accuracy. Before any model sees a timeline, a three-phase procedure replaces every named entity and shifts every date by a per-event constant, turning a real arc into a counterfactual social world -- structurally identical to what happened, but stripped of the surface labels a model could match against pre-training memory. On five heterogeneous events and 125 prediction points in Chinese and English editions, the strongest of six frontier LLMs reaches only 75.0 out of 100, against a trivial anchor of 50. The two axes come apart: a model can be calibration-strong but time-weak, or the reverse. Three agent frameworks built on a shared base model fail to improve on that base, and two model-free heuristics trail every LLM. Per-event gaps reach 21.4 points on a single axis, which is our main argument for evaluating on several events rather than one. All anonymized timelines, question banks, ground truth, and scoring code are released.

Benchmarks that measure the forecasting ability of large language models are almost always retrospective: the event has happened, the answer is somewhere on the Web, and the evaluation must defend itself against memorisation. We report the opposite design. Over the 39 days of the 2026 FIFA World Cup, six frontier LLMs -- all with extended thinking and native server-side web search -- were asked before every kickoff, one match at a time, to fill in a seven-market prediction card for all 104 matches, plus 12 group winners and a pre-tournament outright pool; no answer existed when the question was asked, so the evaluation is leakage-free by construction rather than by filtering, and the frozen archive holds 4,494 scored predictions. What the tournament establishes is a set of behaviours the six systems share. On match outcome they average 63.9%, level with backing the bookmaker's favourite -- which is in fact what they usually do. They agree with one another far more often than they are right, so a majority vote adds nothing. They under-commit to draws and to goals, and crowd their scoreline picks onto a single prototypical result. Accuracy tracks how lopsided a fixture is rather than how much is known about it: it collapses in the closest ties, where the dossiers are richest, while questions about the tournament as a whole are answered well. On this task the current generation of frontier systems is not sharply differentiated: the standings hold up at the top and the bottom across the run and churn in the middle, and the margins stay narrow throughout. The briefing dossiers, fixtures and official results are released as a benchmark, together with the scoring code.

Tool-Integrated Reasoning (TIR) enables LLMs to solve complex tasks through iterative tool interactions. However, existing reinforcement learning methods often rely on trajectory-level supervision, limiting fine-grained credit assignment in long-horizon TIR scenarios. On-policy self-distillation offers denser signals through teacher branches with privileged context, but existing approaches typically derive such context from ground-truth answers or retrieved skills, which may not reflect the states actually visited by the agent. Moreover, token-level supervision fails to capture the turn-level structure of tool interactions. To address this, we propose TurnSight, a turn-level hindsight self-distillation framework that derives supervision directly from execution-conditioned hindsight. It then constructs multiple hindsight views with different lookahead horizons and selects reliable supervision through cross-horizon directional agreement. Finally, the selected hindsight signal is normalized across sibling rollouts and used to adaptively modulate RL advantages while preserving their original optimization direction. Extensive experiments on three benchmarks demonstrate the effectiveness of TurnSight. Our codes are available at https://github.com/quchangle1/TurnSight.

Recursive self-improvement requires agents to turn accumulated experience into better future behavior. Personal AI agents offer a concrete setting for studying this capability because they retain preferences, task histories, tool routines, and learned skills across sessions. Yet whether retained experience actually improves them over time has not been systematically tested. We introduce PAST-Bench, a benchmark designed to isolate this question. Each agent runs through ordered sequences of fresh-session tasks under matched conditions that turn retained experience on and off. It spans 26 scenarios and 204 episodes across memory, procedural reuse, information gathering, and update. We report both later-task gains and whether those gains follow the intended save, retrieve, and update pathway. Across seven base models and four agent frameworks, improvement is real but uneven across capabilities. Agents with the same headline gain can differ markedly in whether that gain is supported by evidence of the intended pathway. Guided by these findings, we develop Hermes+, which extends Hermes with five targeted interventions across stages of the agent loop. Hermes+ raises the average gain from retained experience and provides clearer pathway evidence, with its strongest improvement on tasks requiring outdated state to be replaced, although the effect remains capability- and model-dependent. Together, PAST-Bench and Hermes+ provide an evaluation and diagnostic foundation for studying how persistent agents can progress from retaining experience to systematically improving through it. Code: https://github.com/Gen-Verse/PAST-Bench

Large language models can solve substantially harder reasoning problems with more inference-time compute. The term "test-time scaling," however, now covers diverse inference algorithms that extend deliberation along a single trajectory, sample completed candidates and aggregate them through voting or verification, or search over unfinished partial states. These algorithms differ in their statistical structure, compute accounting, and failure modes. Treating these procedures as interchangeable under a single scalar "budget," or reporting accuracy without the inference protocol that produced it, makes results difficult to compare across studies. We develop a systematic account of test-time scaling along three axes. First, we formalize test-time scaling as budgeted inference over the implicit prefix tree of an autoregressive model and distinguish three structural regimes: single-trajectory sequential scaling, leaf-level scaling with terminal reduction, and prefix-level scaling. Second, we treat the evaluated object as the entire inference system and develop evaluation principles that separate end-to-end system performance from candidate-bank diagnostics. We introduce an evaluation profile whose coordinates and simple functionals recover or bound common repeated-sampling metrics, and prescribe protocol-matched reporting of compute and uncertainty. Third, we specify reproducibility requirements for inference protocols, distinguishing exact replay from distributional reproducibility and identifying the artifacts needed to support each. We also organize the open-weight reasoning ecosystem by model-side and interface mechanisms, apply these principles to broad-knowledge, symbolic-reasoning, and competition-mathematics benchmarks, and assemble over 2 billion full reasoning traces for release with progressively richer verifier and token-level signals.

We identify a previously overlooked failure mode of ALiBi positional encoding: its linear bias scaling underflows floating-point precision, which zeroes out a large fraction of attention weights and renders the affected attention heads partially blind. We analyze this failure mode, characterize its impact, and examine four mitigation strategies. We further demonstrate its occurrence in state-of-the-art pretrained models based on ALiBi. Comprehensive pretraining experiments with 148M-parameter decoder models help us to disentangle its effects from out-of-context degradation. We find that ALiBi's failure mode can substantially impair token retrieval while having only a minor effect on standard decoder benchmarks. We propose four training-time mitigation strategies and evaluate them individually and in combinations, finding that log-scaled distances yield the most consistent improvements in passkey retrieval. Despite this problem, default ALiBi slopes remain a surprisingly strong baseline, particularly for needle-in-a-haystack retrieval. Based on these findings we provide concrete recommendations on how to train models with ALiBi.

Synthetic histopathology image generation has emerged as an approach that may address data scarcity in computational pathology, yet current evaluation methodologies may not fully assess synthetic data quality for medical applications. This work investigates and addresses limitations in existing evaluation metrics, investigating an approach for assessing synthetic histopathology image quality through domain-specific metrics and downstream task validation. We show that conventional synthetic data evaluation metrics such as Frechet Inception Distance (FID) and Inception Score (IS) may have limitations when applied to histopathology images due to their reliance on ImageNet-pretrained feature extractors. To address these limitations, we propose for consideration modified FID and IS approaches utilizing foundation models pretrained on digital pathology datasets, supplemented by precision-recall based metrics as part of an additional quality assessment. Using conditional denoising diffusion models trained on four benchmark datasets, with a two-step training approach, we generated synthetic datasets with systematically varied quality characteristics. We also measured the correlation between the synthetic data quality metrics with downstream nuclei segmentation performance using common metrics including the aggregated Jaccard index (AJI+) and the Dice coefficient. The study results suggest that pathology-specific metrics may provide improved discriminative power. Specifically, the modified Inception Score indicates higher correlation with downstream task performance (r=0.6096 with AJI+, p=0.0122), compared to the original IS (r=0.0708, p=0.7944). Our observations indicate that increasing the variety of generated training data has a higher positive correlation with segmentation model performance than improving the visual fidelity of individual generated images.

We present string2string Studio, an interactive in-browser platform for string-to-string analysis across natural language processing, computational biology, and the digital humanities. The system integrates six main modules (alignment, distance, similarity, search, generation metrics, and BLAST homology search), operating at character, word, token, line, and residue levels. Its C++-based algorithms compile to WebAssembly, so core operations run locally by default without any installation or data upload. The interface reports scores with their "evidence" (alignments, edit paths, metric matches, search hits, and homology traces), making methods inspectable, debuggable, and comparable on shared inputs. Internal benchmarks show speedups of up to 2,500x over the Python predecessor, faster global/local alignment than a general-purpose native C aligner, and exact agreement with independent references under declared settings. For homology search, the scoped client-side blastn path closely matches NCBI BLAST+ rankings and statistics under matched parameters. A curated showcase and Learn mode present canonical algorithms and metrics as reusable demonstrations. string2string Studio is open-source and freely available at string2string.org.

Optimizing compilers miss profitable transformations when their enabling semantics are absent from the analyzed program representation. We ask whether large language models (LLMs) can recover such semantics from heterogeneous C/C++ context and realize them as validated, contract-preserving artifacts. We introduce SeGaBench, an executable benchmark containing 100 synthetic and 20 source-backed cases spanning low-level assumptions, data-structure invariants, and high-level semantic lifting. Each case includes hidden enabling semantics, an oracle artifact, correctness and semantic validators, and a reproducible performance protocol. We evaluate five LLMs using five independent responses per case. The strongest model produces correct artifacts in 94.8% of responses, achieves at least 1.05x speedup in 83.3%, and obtains a performance success on 93.3% of cases. Nevertheless, correct artifacts often close only part of the oracle gap. These results show that LLMs can complement compiler analysis as speculative semantic proposers, provided that their artifacts are validated and evaluated.

We introduce Video-DeepResearch (Video-DR), extending multimodal agents from static images to continuous video streams, a setting that demands dense spatiotemporal grounding coupled with open-web exploration. Preliminary evaluations reveal two critical bottlenecks in current models: (1) modality bias, where agents bypass visual tools in favor of textual search, and (2) parametric knowledge leakage, where models rely on internal memory rather than genuine tool-augmented execution. To address these challenges, we propose Video-DR, featuring a decoupled perception-exploration pipeline with stage-wise tool unlocking that compels exhaustive cross-frame visual grounding prior to web retrieval. Our framework adopts a two-stage training recipe: supervised fine-tuning followed by Group Relative Policy Optimization (GRPO), enabling autonomous exploration that breaks the imitation-learning ceiling. Furthermore, we curate Video-DR-Bench, a human-AI collaborative benchmark comprising 200 complex, multi-hop VQA instances. Empirical results demonstrate that our Video-DeepResearch-35B-A3B establishes a new state-of-the-art of 64.0% average accuracy, surpassing proprietary Claude-4.5-Sonnet (59.0%) by 5.0 points and significantly outperforming GPT-5 (52.5%) and Gemini 2.5 Pro (57.5%). The 30B-A3B variant achieves 59.3%, competitive with Claude-4.5-Sonnet and demonstrating the effectiveness of our training paradigm even at compact scale. Code: https://github.com/Osilly/Vision-DeepResearch.

Stochastic single shooting trajectory optimization methods such as Model Predictive Path Integral control (MPPI) have been widely adopted in robotics due to their ability to reason about probabilistic dynamics and provide solutions where model gradients are noisy, costly to evaluate, or unavailable. However, satisfaction of terminal constraints when shooting over long action sequences is often sample inefficient, requiring a large number of iterations for convergence. In this paper, we present a stochastic multiple shooting method that optimizes short control action sequences connected via local feedback policies to improve sample efficiency and convergence to a terminal set. Additionally, we show that we are able to synthesize approximate system Jacobians purely from rollouts, making the method suitable for model-based reinforcement learning with black-box dynamics. We demonstrate the algorithm has improved sample efficiency and terminal set convergence for three nonlinear, underactuated optimization problems: a classic cartpole swingup task with analytical dynamics, a cartpole swingup task with learned neural network dynamics, and a VTOL quadplane performing a high angle-of-attack, precision post-stall landing maneuver.

On-policy training has emerged as a powerful post-training paradigm for improving the reasoning capabilities of large language models, and is often enhanced by golden trajectories from stronger expert models. However, when the expert fails on harder problems, existing trajectory-guided methods lose their main source of supervision, and these failed trajectories are typically discarded as negative samples. We argue that such failures, which we call Golden Negative Trajectories, can still provide valuable reasoning signals when treated not as demonstrations to imitate, but as flawed trajectories to reflect upon. We identify a Reflection Advantage: for hard problems, reflecting on a flawed trajectory can be easier and more effective than solving the problem directly from scratch. Motivated by this, we propose ReflectRL, a lightweight plug-and-play framework that learns from Golden Negative Trajectories during on-policy training. ReflectRL first uses these trajectories to elicit Reflective Reasoning, then applies Reflective-to-Direct Policy Transition to transfer the acquired reasoning behavior back to Direct Reasoning. Experiments across 9 benchmarks, 4 LLM backbones, and 4 on-policy training methods show that ReflectRL consistently improves reasoning performance with minimal overhead.

Human input reaches language models by typing or speaking, and each channel leaves a distinct signature: orthographic noise for keyboards; for voice, disfluency from conventional transcription and restructuring from AI-backed dictation tools. How do they impact an LLM's performance? In this paper we present HIVE (Human Input-Variation Engine), a suite of voice transcription perturbations and QWERTY keyboard perturbations. We use HIVE to evaluate how robust models are to these perturbations. We present seven findings. (i) Voice transcription perturbations lower accuracy across every instruction-tuned model we test, and it is the structure of the transcription rather than its fillers that carries the cost. (ii) QWERTY keyboard perturbations cost less, and a model absorbs a lot of them before accuracy falls away. (iii) Both trace back to one cause, how many of the question's tokens survive the perturbation: destroying a token is what hurts, while adding new ones alongside it costs little. (iv) The gap between the two channels appears only where the answer must be constructed or deduced; on multiple choice there is none. (v) The harm does not solely come from test-set contamination. (vi) It cannot be trained away with lightweight adaptation. (vii) A thinking budget recovers the keyboard channel almost entirely but leaves the spoken registers untouched, and compressed speech is worse with it.

As autonomous agents powered by foundation models are increasingly integrated into social and economic systems, understanding the principles governing their collective behavior is essential for ensuring safety and cooperation. Classical game theory, the dominant framework for modeling rational interaction, is built upon the assumption of `decoupled agency,' where agents treat their own decision-making as independent of the environment and other actors. Modern AI agents, however, jointly predict their own future actions alongside external observations. Here, we report a striking finding: when interacting in stylized social dilemmas, foundation model agents engaging in optimal planning consistently converge to stable cooperation, directly contradicting classical game-theoretic predictions of mutual defection. To understand this phenomenon, we introduce the `embedded Bayesian agent,' a theoretical model for foundation model agents. By shifting from decoupled to embedded agency, these agents model themselves as part of the universe they inhabit, maintaining epistemic uncertainty about their own decision-making algorithms. We show that by inferring whether others are behaviorally similar, an embedded agent treats its own deliberation during planning as evidence: a decision to cooperate predicts a similar decision by a similar partner. We formalize this mechanism of similarity inference through the `embedded equilibrium,' a novel solution concept replacing the Nash equilibrium to provide a foundational game theory for the social behavior of modern AI agents.

Large language models (LLMs) are increasingly used to provide conversational practice for English-as-a-second-language (ESL) learners. Effective ESL tutoring, however, requires more than fluent response generation: a tutor must select an appropriate pedagogical action based on learner behavior and dialogue context. Human-tutoring research offers principles for adaptive support, but they are often task-specific and remain insufficiently integrated into LLM-based ESL tutor training and evaluation. We present TACT (Taxonomy-Aligned Conversational Tutor), a human-grounded framework for post-training and evaluating pedagogically adaptive ESL tutors. Drawing on established literature, we develop two complementary taxonomies: the Tutor-Strategy Taxonomy with 13 tutor response strategies and the Student-Move Taxonomy characterizing learner behavior by move type and status. Using these taxonomies, we construct TACTCorpus, which enriches 260 authentic teacher-student conversations with 32,379 annotations and quality-controlled augmented training data. We then post-train Qwen3.5-4B through supervised fine-tuning followed by taxonomy-aligned Group Relative Policy Optimization, producing TACTutor and optimizing it for scaffolding quality rather than reference imitation alone. On TACTBench, a strategy-balanced diagnostic benchmark comprising 78 authentic tutoring contexts, TACTutor improves over its backbone by 20.30% and outperforms all evaluated proprietary baselines under the same protocol, while maintaining backbone performance on established external educational benchmarks; in a blinded study with 50 learners, it also receives the highest overall mean rating among the evaluated tutors. We release the data, benchmark, and model weights, providing an open foundation for developing pedagogically adaptive ESL tutors.

Bimanual manipulation policies trained with imitation learning are typically evaluated on workstation or datacenter-class GPUs, leaving the cost of deploying them on embedded hardware largely uncharacterized. We present a bimanual SO-101 system running entirely on an NVIDIA Jetson Orin Nano Super (8 GB), the entry-level tier of NVIDIA's embedded line, using a desktop GPU (RTX 3070) only for offline training, evaluated on pick-and-place of a deformable beanbag. First, we build a GStreamer capture pipeline backed by NVMM buffers that removes redundant host-device copies from three-camera sensing. Contrary to expectation, the conventional path fit the memory budget and dropped no frames; what zero-copy sensing recovers is CPU headroom (peak single-core utilization 98.0% to 77.0%) and worst-case latency (117.31 ms to 101.52 ms). Second, we train ACT and Diffusion Policy on identical demonstrations, each at its own reference budget (100k gradient steps for ACT, 200k for Diffusion Policy). ACT converges to a task-competent policy (19/20 trials) while Diffusion Policy does not converge to a usable one (0/10) even at twice the step count, which we attribute to differing convergence costs rather than an accuracy ceiling. Third, we convert ACT to TensorRT. FP16 reduces mean inference latency from 114.02 ms to 17.93 ms (6.4x) and INT8 to 12.65 ms (9.0x), with task success preserved at all three precisions (19/20, 18/20, 19/20). We report two findings not previously documented for ACT: TensorRT's general-purpose INT8 calibration quantizes the ResNet18 backbone but accepts zero of 145 transformer layers, explaining INT8's negligible size reduction over FP16 (0.9%) despite a further 28% latency gain; and the need for quantization is conditional on ACT's action-chunking configuration, feasible in full precision at n_action_steps = 100 but not at the per-step re-prediction temporal ensembling requires.

Pre-pretraining language models (LMs) on symbolic data can accelerate and improve natural language acquisition. However, existing pre-pretraining tasks, such as Dyck and procedural algorithms, rely on narrow primitives that fail to capture the expressive capacity of natural language. Moreover, prior studies remain restricted to relatively small token budgets, offering limited insight into skill emergence and representational dynamics. To address these limitations, we propose logic pre-pretraining (Logic-PPT) as a principled initialization strategy, leveraging formal derivations to impart richer structural and linguistic biases. Formal derivations require abstract mechanisms that are central to natural language, simultaneously binding variables, connecting quantifiers and relational dependencies, and composing predicate-argument structures over long contexts. Scaling our evaluation to a 100B-token regime, logic pre-pretraining substantially accelerates skill acquisition in LMs, achieving 80\% accuracy on linguistic tasks with 36B fewer tokens than standard initialization, and outperforming alternative pre-pretraining baselines. Mechanistically, formal derivations induce persistent structural reorganization, distinctively characterized by a lower-rank, spectrally concentrated representation space. Crucially, we show that this internal geometry enables improved model compressibility via pruning, matching the dense baseline performance even at $\approx$33\% sparsity.

Tensor cross-concentrated sampling (t-CCS) bridges entrywise sampling and t-CUR slice-wise sampling by observing entries only within selected horizontal and lateral slices. Existing t-CCS completion methods, however, assume that the observations are free of gross corruption. In this work, we study robust recovery of a third-order low-tubal-rank tensor from partial t-CCS observations contaminated by sparse, arbitrarily large outliers. We propose Robust Iterative t-CUR (R-ItCUR), a tensor-native algorithm that partitions the sampled tensor cross into two exterior blocks and an intersection block, applies adaptive blockwise Welsch correction for outlier suppression, and updates the low-rank component through projected blockwise gradient descent. By operating directly on the sampled cross, R-ItCUR avoids reconstructing the full tensor throughout the iterations, resulting in substantial memory and computational savings. Experiments on synthetic tensors, cardiac MRI data, and three-dimensional seismic data demonstrate accurate recovery and strong robustness to sparse gross corruptions. The results further highlight the importance of explicitly exploiting the cross-concentrated sampling structure in robust tensor completion.

Engineered Skeletal Muscle Tissues (ESMs) have become a key structure for biomedical disease modeling and pharmacological screening, yet their functional characterization often relies on simplistic metrics like peak force, discarding critical kinetic information. This is partially due to the high level of mathematical complexity which mechanistic models introduce to capture these dynamics. Hence, exactly the complexity prevents scalable application and widespread adaptation in the field. Here we present a Physics-Flavored Neural Network (PFNN) that automates the kinetic phenotyping of ESMs. Our architecture integrates a stretched-exponential physical model into a CNN-Transformer, enabling the extraction of physically meaningful parameters directly from force-time profiles. To address the scarcity of labeled biological data, we employ a hybrid training paradigm: the model develops a "physical intuition" on synthetic data before undergoing unsupervised self-alignment on unlabeled real-world measurements. Our results demonstrate that this physics-flavored approach achieves high-fidelity parameterization across diverse contractile phenotypes and cell lines, including Duchenne Muscular Dystrophy models. Our scalable, self-improving pipeline bridges the gap between idealized biophysics and noisy \emph{in vitro} data, providing a robust tool for high-throughput biophysical research.

When will robots have their ChatGPT moment? Such a breakthrough requires a general-purpose robot that can handle unfamiliar tasks in unfamiliar environments, remain controllable over long interactions, and learn from experience. Today's embodied systems largely follow an end-to-end observation-to-action path. Despite rapid progress, they remain far from this goal: their generalization depends heavily on the coverage of robot training data, while long task execution remains difficult to control and inspect. To realize this goal, we introduce the Embodied Task Agent (ETA), a new paradigm for extending digital agents into the physical world, and release OpenETA as its open-source implementation. ETA centers the robot around a Planner that chooses one Tool call at a time, an Interface that controls execution, and a World that returns the result and a fresh observation. This loop allows the agent to verify outcomes, adapt its plan, and turn successful and failed interactions into reusable experience. OpenETA provides replaceable Planners, composable Tools and Skills, auditable memory, replayable trajectories, and common interfaces for simulation and real robots. For Codex, OpenETA can operate as a lightweight plugin that exposes only observe, mark_point, and move_to.

Trajectory inference is a fundamental problem in many scientific domains: given a collection of unpaired snapshots of observations at discrete time points, the goal is to generate smooth trajectories that best resemble and interpolate the data. Existing algorithms exhibit computational challenges: they either rely on preprocessing subroutines to enforce smoothness or on simulation-based training objectives, both of which can be expensive. In order to overcome these limitations, we propose a new algorithm called Acceleration Matching (\texttt{AM}). Our approach consists of lifting the original interpolation problem to phase space and then regressing onto an explicit conditional acceleration field that induces random, smooth trajectories that agree with the prescribed marginals. Importantly, our resulting training algorithm only requires positional data, avoids trajectory simulation during training, and is devoid of expensive preprocessing. We provide ample numerical evidence suggesting that \texttt{AM} is competitive with or superior to existing algorithms on several benchmark problems from the existing literature.

As AI systems are deployed across increasingly diverse social contexts, alignment can no longer be framed as the optimization of a single, unified set of values. Instead, systems must be able to recognize, represent, and respond to multiple legitimate perspectives. This has led to growing interest in pluralistic alignment, which seeks to move beyond one-size-fits-all models of appropriate behaviour. However, current approaches often lack a clear account of how values are socially organized, contested, and coordinated in practice. In this paper, we argue that social theory provides essential conceptual and design resources for addressing these challenges. Drawing on established traditions in sociology, we show how perspectives can be understood as structured by roles, shaped through interaction, and distributed across fields of power and expertise. We translate these insights into concrete implications for AI system design, including role-based representations, structured coordination among perspectives, and context-sensitive evaluation. For agentic systems, this requires aligning not only final outputs, but also the role activations, deliberative traces, aggregation rules, and feedback loops through which those outputs are produced. Our contribution is to reposition pluralistic alignment as a problem of socially grounded coordination rather than output diversification. We outline a design space for systems that engage multiple perspectives in structured and accountable ways, and we identify directions for future work to implement and empirically evaluate these approaches in real-world settings.

The automatic structural analysis of legal texts is a cornerstone of legal technology, yet the extraction of their logical components remains a significant challenge. In this paper, we introduce the task of identifying and segmenting legal conditions (Tatbestand) and legal consequences (Rechtsfolge) within German statutory texts. To support this task, we present ANNOTARES (Annotations of Tatbestand-Rechtsfolge Sequences), a novel dataset comprising German law texts with span-level annotations. Spanning three distinct legal codes, the dataset is designed to evaluate both domain-specific performance and cross-statute generalizability. We benchmark diverse architectural approaches: a rule-based baseline, CRFs, BiLSTMs, BiLSTM-CRF, and modern Transformer-based models, including BERT variants and LLM-based methods. Our results demonstrate that BERT and LLM-based models achieve superior performance in capturing the complex syntactic structures of legal language. We release our dataset to facilitate further research in automated legal reasoning.

We study linear representations of temporal horizon in the large language model Qwen3-32B and use them to change the model's time-related preferences, recommendations, and capabilities. We train contrastive linear probes on teacher-forced temporal-choice answers to find a short-term versus long-term direction in the model's residual stream, and evaluate contrastive activation-addition steering on a held-out binary temporal-choice task, an out-of-distribution monetary intertemporal-choice task, and a TravelPlanner capability benchmark. The central result is that temporal-horizon directions can be identified with simple contrastive linear probes and then used for steering to induce large, bidirectional preference changes. On an out-of-distribution monetary choice task that varies reward size and delay, steering strongly shifts the model's indifference threshold between smaller-sooner and larger-later rewards in both directions. We further show improvements on a planning-related capability metric under moderate temporal steering. These results suggest that model intertemporal preferences are measurable and steerable, which is relevant for AI systems that give advice involving delayed costs and benefits, and for safety questions about long-horizon planning.

A clinically useful chest X-ray system must go beyond fluent report generation: it should classify findings with tunable decision thresholds, localize them spatially, and derive the anatomical measurements upon which many diagnoses depend. Today's Vision-Language Models (VLMs) treat these as separate problems, if they address them at all, leaving a gap between what radiologists need and what generative models provide. We introduce CARE-X, a chest X-ray VLM that narrows this gap by unifying auxiliary discriminative supervision with reward-aligned generation. CARE-X augments its generative backbone with focal-loss classification and composite-loss grounding heads, co-trained alongside the language-modeling objective. This auxiliary supervision produces discriminative diagnostic predictions with tunable decision thresholds and precise spatial localization while also improving report quality, providing evidence that structured prediction and generation reinforce one another. Building on this foundation, Decoupled Clip and Dynamic Sampling Policy Optimization (DAPO) leverages task-specific reward signals for report generation, visual question answering (VQA), and spatial grounding, directly optimizing the clinical quality metrics that matter in practice. The result is state-of-the-art performance on the majority of metrics across four report-generation benchmarks, 94.0% VQA accuracy on ReXVQA (+6.0 pp over the next-best baseline), and generative spatial decoding that reaches near parity with dedicated detection heads. Separately, to address measurement-dependent diagnoses, we couple Qwen3-VL-4B-Instruct with native tool-calling capabilities for invoking deterministic measurement tools, while retaining full visual access to the image. This hybrid inference yields +43.6 pp average F1 over perception-only baselines across five measurement-dependent conditions.

In-the-wild Bengali scene text recognition is largely unmeasured: existing resources target handwritten documents or constrained sign-board parsing, report only aggregate edit-distance metrics, and evaluate either conventional OCR or VLMs, never both on the same in-the-wild data. To address this gap, we introduce BANGLAWILD, a benchmark of 2,535 Bengali scene text images, each paired with a verbatim gold transcription, two categorical axes, four diagnostic attributes, and an orthographically standard form where the in-image text deviates from canonical spelling. We evaluate fifteen VLMs and three conventional OCR systems under three prompting strategies, fine-tune 6 open-source models with LoRA, and complement edit-distance metrics with an LLM-as-a-Judge evaluation. Our results reveal a persistent gap in which larger models within the same family do not outperform smaller ones. Our fifteen-class error taxonomy shows that visual mis-recognition accounts for ~60% of errors in the strongest systems, while conjunct-related errors contribute under 2%, challenging a long-standing assumption in Bengali OCR research; the same visual dominant profile also holds across architectures, including the one conventional baseline that reads Bengali reliably. Prompt language mainly affects cross-script drift and LoRA reduces catastrophic failures in weak models without lifting the ceiling on already competent ones. Code and data will be publicly released.

Geospatial reasoning, i.e., computing distances, containment, and other spatial relations over real-world entities, is central to navigation and logistics, yet large language models (LLMs) struggle with the required geometric and topological computation despite storing considerable geographic knowledge. Existing benchmarks localize these failures only partially: they are synthetic or smallscale, largely monolingual, and offer limited control over geographic coverage. We introduce MultiGlobeQA, a multilingual benchmark of 46,060 question-answer pairs spanning 14 spatial-function families and 15 answer formats, with execution-based ground truth over three knowledge graphs. It covers 201 countries and territories via income- and density-stratified sampling, with parallel questions in English and 16 additional high- and low-resource languages. Across parametric, reasoning, and agentic settings, LLMs collapse on tasks requiring grid indexing and shape computation, while topological relations and directions fare best. Retrieval and tool use yield considerable gains, yet performance plateaus below two thirds even when gold facts are supplied, indicating that computation, not access to knowledge, is the bottleneck. Models also underperform on low-income regions, a gap that gold facts widen rather than close.

Synthetic power-grid scenarios are essential for planning, resilience assessment, contingency analysis, and data-driven power-system applications. Recent synthetic grid generation methods have improved structural realism and operational feasibility by incorporating engineering knowledge through post-generation validation, optimization, or physics-aware generation. However, generated scenarios may still exhibit low AC feasibility and robustness, limiting their practical value for downstream power-system studies. This paper proposes a feasibility-aware distribution-learning framework that learns the AC-operable joint distribution of network topology, branch electrical parameters, and time-varying load profiles. Instead of enforcing feasibility after generation, the proposed framework incorporates AC power-flow convergence and operational constraints into hierarchical diffusion-based distribution learning. This enables the generator itself to produce operationally feasible grid scenarios through efficient diffusion sampling. The hierarchical architecture decomposes the high-dimensional generation task into three engineering-motivated stages: topology and bus-attribute generation, branch-parameter generation conditioned on the generated structure, and load-profile generation conditioned on both network structure and electrical characteristics. Experiments on benchmark systems demonstrate that the proposed framework significantly improves operational feasibility and contingency robustness while maintaining strong statistical fidelity and eliminating optimization-based post-processing.

Modern agent frameworks equip large language models with external skill libraries to solve complex tasks. However, it remains unclear whether these systems can effectively evolve their skills and whether the resulting skills improve task-solving capabilities. To bridge this gap, we introduce ContinualSkillBench, a dynamic evaluation framework for in-context continual skill learning. It covers five representative domains, each containing 100 interconnected subtasks ordered by increasing difficulty and opportunities for cross-task skill reuse. Our experiments show that sequential execution generally improves performance, but the gains vary substantially across models and domains. Moreover, in-context learning performs comparably to explicit skill maintenance on average, suggesting that much of the improvement arises from adaptation to prior context and feedback rather than reusable skill abstraction alone. Explicit skills nevertheless provide selective benefits for tasks requiring reusable procedures or precise outputs. We further find that less capable models tend to accumulate larger, more fragmented collections of task-specific skills. These findings show that current in-context skill evolution mechanisms can support continual adaptation, but still struggle to consistently consolidate experience into robust and transferable skills.

Human-in-the-loop reinforcement learning (HIL-RL) enables robots to learn contact-rich manipulation from limited real-world interaction, but deployment exposes three coupled limitations: static visual reward models fail under scene changes; independently sampled actions cause temporally inconsistent motion; and vision-based policies remain sensitive to appearance shifts. We present EvoHIL, a unified framework that adapts the reward model, action generator, and visual do main within a staged human-in-the-loop learning process. First, self-evolving reward (SER) adapts the success classifier from human-confirmed positives and provisional weak negatives. Second, Action Flow Stabilization (AFS) generates temporally coherent action chunks through flow matching, grounding policy updates in executed action prefixes and demonstrated behavior. Third, retention-aware offline fine-tuning replays relit interaction data while anchoring the AFS actor-critic to prior behavior, adapting the visual domain without additional robot interaction. Across six manipulation tasks on Franka FR3 and SO-101 arms under a controlled lighting shift, EvoHIL improves task success, agreement with human-confirmation labels, motion smoothness, and completion time relative to human-in-the-loop and imitation baselines.Project page: https://anonymous4366.github.io/EvoHIL/

Causal Discovery (CD) from observational data faces two fundamental challenges. First, purely statistical methods often lack the power to resolve structural ambiguities in low-sample regimes. Second, although LLM-assisted hybrid approaches improve structure recovery through semantic reasoning, the influence of that reasoning on individual edge decisions remains largely opaque. Consequently, existing hybrid methods fail to satisfy a fundamental requirement: explaining why a particular edge is included or excluded in the learned directed acyclic graph (DAG). This is critical in real-world applications, where no ground-truth DAG exists and every structural decision must be independently justified. We formalize this requirement as decision traceability, requiring every inferred edge to be supported by auditable statistical evidence, Markov Blanket consistency, or explicit domain reasoning. We propose GENESIS, an explainable hybrid CD framework that decomposes graph construction into interpretable decision points. GENESIS first identifies and scores three-node structural motifs, including chains, forks, and colliders, to establish transparent structural priors, then progressively refines the graph by integrating these priors with observational evidence, invoking domain knowledge only when statistical evidence is insufficient. By design, every edge decision is resolved through an auditable source of evidence. Experiments show that GENESIS achieves 100% decision traceability across all settings, establishing explainability as a first-class objective in causal discovery. Despite this additional requirement, GENESIS consistently outperforms purely statistical CD methods on the majority of benchmark datasets across all sample regimes in terms of Structural Hamming Distance (SHD), while achieving performance comparable to state-of-the-art LLM-assisted approaches.

This white paper presents ADMITBench, a reference framework for evaluating industrial LLM advisories at the level of the proposed action. The framework implements a versioned, safety-governed evaluation contract that checks whether a recommendation is supported by the available evidence, permitted under the stated authority and procedure, and acceptable under the plant-specific consequence checks encoded in the selected evaluation profile. In this report, \emph{safety-governed} means that eligibility is determined through explicit, non-compensatory checks derived from a versioned plant profile; it does not mean that the evaluator, model, or plant has been safety-certified. Release 0.1.0 is a public reference implementation for technical and research evaluation, not an authorisation for physical execution.

We introduce SciRet, a compute-aware empirical study of retrieval-augmented generation for scientific question answering over CORD-19. Rather than proposing a new model, we evaluate a fixed scientific RAG pipeline across three corpus scales: 1,034 chunks (1K papers), 5,160 chunks (5K papers), and 15,480 chunks (15K papers). The pipeline combines sentence-window chunking, BM25, BGE-M3 dense retrieval, reciprocal rank fusion, optional cross-encoder reranking, and grounded answer generation. Across these settings, hybrid retrieval is more robust than either sparse-only or dense-only retrieval in our setting, reaching Recall@10 of 1.000 at 1K and 15K. In contrast, an MS MARCO-trained cross-encoder reranker reduces precision on the scientific corpus, suggesting that domain mismatch can outweigh the benefits of stronger query-passage interaction. Generation faithfulness measured with RAGAS increases with corpus scale in our setup. Retrieval evaluation uses pseudo-relevance labels derived from the hybrid system, so we treat the results as controlled comparative evidence rather than a benchmark claim. We release code, indexes, and evaluation outputs to support replication and follow-up studies.

Large language models (LLMs) pose challenges to academic integrity and peer review. Yet generative plagiarism detection remains an underexplored and largely unresolved challenge. Prior work on LLM-generated-text detection targets AI involvement, which may be permissible, rather than source reuse, while similarity-based methods struggle after extensive rewriting and multi-source synthesis. Motivated by the description-length view of probabilistic prediction, in which relevant side information can reduce a target sequence's code length, we introduce Source-Conditioned Description-Length Gain (SCDG), a directional, training-free framework that contrasts a frozen language model's description length of a suspicious document $P$ with and without a candidate source $S$. This contrast yields token-level log-likelihood gains that measure the incremental predictive evidence supplied by $S$. We evaluate SCDG on the PAN at CLEF benchmarks for generative plagiarism. On a PAN 2025-derived pairwise benchmark, SCDG achieves 0.92 Precision, 0.97 Recall, and 0.94 F1, outperforming all baselines; on PAN 2026's multi-source retrieval task, it reaches 0.83 nDCG@10 and 0.96 Recall@100, surpassing all baselines. On a same-topic, same-event Multi-News test, the calibrated gain-distribution SCDG classifier predicts source reuse for only $0.125\%$ of pairs, supporting robustness to topical overlap under this evaluation protocol. These results establish SCDG as a unified and token-decomposable signal for source-specific content reuse under extensive transformation.

Transformer models have revolutionized natural language processing (NLP), and text-based molecular representations like SMILES have successfully extended these architectures to chemistry. However, domain-adaptive pre-training often causes models to overfit to chemical syntax, catastrophically forgetting their foundational semantic capabilities. To address this challenge, we introduce CheMatE, a chemistry-oriented embedding model that jointly captures molecular structure and domain-specific natural language within the same representation space. Built on a ModernBERT backbone, CheMatE learns bi-semantic representations through a two-stage training procedure: continued masked language modeling (MLM) followed by a Matryoshka contrastive learning stage via Multiple Negative Ranking Loss (MNRL). First, we train the model using MLM on a novel, large-scale corpus of SMILES-annotated, long-context scientific documents that were constructed and curated from FineWeb and ChemPile (comprising 10.4B and 11.5B tokens, respectively). Subsequently, the model undergoes contrastive learning using a synthetic dataset of SMILES-text pairs algorithmically derived from our original training corpus. This design exposes the model to SMILES-enriched scientific literature, enabling bi-semantic understanding. We evaluate CheMatE across a range of downstream tasks covering molecular property prediction and scientific language understanding. Our results demonstrate that coupling our custom-curated datasets with this sequential training strategy yields robust, highly transferable representations. By effectively unifying structural and contextual signals within a single text-based framework, CheMatE achieves competitive performance across both specialized chemistry models and general-purpose language model baselines.

This paper proposes FedCritic-MIMO, a communication-efficient serverless federated multi-agent reinforcement learning framework for AI-native resource control across independently deployable cell-level controllers in open and disaggregated 6G RANs. Controllers share no trainer, retain local actors and personalized critic components, and exchange only compatible shared critic parameters. FedCritic-MIMO targets reuse-$1$ multi-cell massive-MIMO OFDMA deployments, where RAN controllers jointly manage user scheduling, per-stream power allocation, beamforming, interference, and long-term QoS with limited inter-controller signaling. Each base station locally executes its actor without centralized training or actor federation, while critic knowledge is exchanged peer-to-peer over an interference-aware graph. It enables this collaboration through wireless-aware event triggering, adaptive layer-wise top-$k$ sparse critic exchange with error feedback, and balanced interference-aware fusion. We establish conditional finite-time stationarity and consensus guarantees for the balanced, compressed peer-to-peer critic recursion under a fixed-policy, frozen-target critic-regression model. In strongly interference-coupled reuse-$1$ simulations, FedCritic-MIMO achieves the best performance-communication tradeoff among heuristic, independent-learning, centralized-training, and communication-ablation baselines. It achieves the highest held-out throughput, improves user-rate distribution and mean SINR, increases QoS satisfaction, and attains the lowest interference cost per delivered bit among learning baselines. It reduces critic-communication overhead by $76\%$ relative to uncompressed distributed critic exchange. These results demonstrate that serverless exchange of compatible shared critic parameters can coordinate RAN controllers without centralized trajectory collection or parameter-server aggregation.

Memory-augmented LLM agents rely on rich context for long-horizon reasoning and acting, yet their memory modules expose a persistent attack surface for malicious records, making the study of memory poisoning threats imperative. However, existing query-only attacks often fail to remain effective in two realistic and prevalent settings: large-scale benign memory pools and active input auditing. Consequently, current approaches fall short when facing the dual challenges of high retrieval competitiveness and rigorous semantic checks. To overcome these limitations, we propose MAFIA, a query-only Memory Attack framework via probing and Factual Injection against Audit, tailored to this extended threat model. Specifically, MAFIA introduces: (1) a placement strategy that ensures retrieval-competitive injection via memory probing, budget allocation, and scheduling; and (2) a payload design that bypasses audits using compact factual cloaks, preserving malicious effects while maintaining high semantic similarity. Extensive evaluations reveal that MAFIA achieves up to a 90.7% attack success rate while suppressing audit detection from a peak of 83.3% to at most 7.4%, exposing critical vulnerabilities across agentic memory systems. Code will be made publicly available at https://github.com/JiamingChen1234/MAFIA.

Training-free speculative decoding drafts by matching an exact suffix of the context against a pool of earlier context. That lookup misses correct drafts already in the pool, most visibly on tool-calling traffic, where a request repeats almost everything but the few values minted for it, and where one rejected token discards the correct continuation behind it. We diagnose the failure position by position across ten benchmarks and find it to be a problem of addressing rather than of coverage: on our densest tool-calling benchmark, about half of what the strongest exact-match drafter misses is present in the pool yet unreachable by exact matching. We therefore propose a second, semantic draft source: the same pool, re-keyed by the hidden state the verifier has already computed at each committed token, together with a merge that lets it ride inside an existing lexical drafter's tree. In three published drafters, at matched pool and budget, it lifts accepted length by 24-29%. Oilbird reaches 4.4x autoregressive decoding speed on API-Bank, against 3.9x for the strongest training-free baseline in our harness and 2.0x for EAGLE-3.

A framework that persists execution state so a run can be interrupted, survive a crash, and continue must decide what a resume means for effects that already fired. Five widely deployed agent workflow frameworks answer differently, none exposes a machine-checkable contract, and behavior violates even the fragments they state. The RESUME CONTRACT states six properties over the persistence API (prefix continuation, effect exactly-once, fork determinism, checkpoint validity, consume-once, recovery determinism), plus fork-intent and liveness obligations. A TLA+ model checks a reference semantics exhaustively, unchanged at scaled bounds (7.4 million states); a 39-cell fault matrix yields the separating models independence requires, and consume-once splits, its consumption clause independent of all six others. A deterministic, LLM-free harness measures them at pinned releases. LangGraph 1.2.9 durably records a second resume value and never consults it, persists schema-invalid state silently, and re-executes durably recorded work after a real SIGKILL: exactly-once across interrupts, at-least-once across crashes, on one API. CrewAI 1.15.2 re-executes completed effect-bearing methods against its written claim; pydantic-graph 1.x cannot resume after a mid-node crash; no two probed frameworks share a conformance profile. Consume-once holds sequentially and fails under concurrent delivery: k processes resuming one parked interrupt fire the gated effect k times, saturation 1.0 in 36 of 40 cells, and the failure crosses hosts. REMIT, a reference sequencer whose Verus-verified recovery core is line-identical to the shipped executable, repairs the fork and validity cells. The cross-process cell is repaired at the read path, and that repair ships: an opt-in gate claims consumption in the shared store, serving one racer and refusing the rest before any node executes.

Geospatial and urban applications increasingly require models to compare heterogeneous evidence across street-view imagery, remote-sensing observations, text descriptions, region proposals, and temporal change cues. However, existing multimodal embedding models and benchmarks are still largely designed and evaluated around general-purpose image-text matching, leaving unclear whether unified embedding space can support heterogeneous geospatial tasks involving spatial relationships, fine-grained semantics, and temporal changes. To address this gap, we make three key contributions. First, we introduce GeoMEB, a large-scale multimodal embedding benchmark that standardizes 45 urban evaluation tasks across retrieval, visual question answering, change detection, classification, and visual grounding, together with training collections comprising 1.32M examples and 286K evaluation queries. Second, we present Geo-Embed, a unified embedding model that adapts a shared vision-language backbone to instruction-conditioned query-target matching over heterogeneous geospatial inputs, including single images, multiple images, text, regions, and masks. On GeoMEB, Geo-Embed achieves the strongest overall performance among representative multimodal embedders, with a 15.3% relative improvement over the strongest baseline. These results motivate future geospatial embedders that organize training and evaluation around explicit query-target relations, including semantic, cross-view, region-level, and temporal correspondence.

Large vision--language models (LVLMs) demonstrate strong multimodal reasoning capabilities but remain prone to hallucination, where model predictions are not grounded in visual evidence. Existing black-box hallucination detection methods estimate uncertainty through a single consistency metric, implicitly assuming that model uncertainty can be adequately characterized by a single measure. However, hallucinations exhibit diverse manifestations of uncertainty across different behavioral probes, making a single measure insufficient to characterize their underlying behavior. We propose \emph{Unique Hallucination Pattern (UHP) Detection}, a fully black-box framework that models hallucination as a structured uncertainty pattern defined by two axes: perturbation modality (image vs.\ text) and logical polarity (a statement vs.\ its negation). Their intersection produces four complementary consistency groups that capture distinct manifestations of model uncertainty, from which both within-group and between-group features are extracted to train a lightweight classifier. Through comprehensive experiments on AMBER and PhD across three LVLMs, UHP Detection consistently outperforms prior black-box and white-box baselines, with improvements of up to $+18.72\%$ AUC-ROC and $+20.07\%$ AUC-PR over the strongest black-box methods. Extensive ablation studies demonstrate that each consistency group contributes complementary information and that their combination forms a structured hallucination pattern. Furthermore, cross-dataset evaluation shows that this learned pattern generalizes across benchmarks, indicating that hallucination behavior reflects a model-specific consistency pattern. \textbf{Code is publicly available at} https://github.com/amirezzati/uhpdet.

We introduce the Unified Neural Variational Measurement of Proficiency (UNVaMP) architecture, a knowledge tracing method that integrates observed student-item interactions with internal memory to produce evolving latent representations of student knowledge. These representations support accurate predictions of future responses while enabling explicit control over the smoothness of estimated learning trajectories. UNVaMP can be configured as either a purely neural model or a hybrid model that predicts responses through an interpretable measurement function over the latent space. We show that a pure neural configuration (UNVaMP-MLP) achieves the strongest predictive performance among compared models on three out of four datasets. Meanwhile, a hybrid configuration (UNVaMP-MIRT, using a 1PL MIRT measurement function) lags only slightly behind UNVaMP-MLP, indicating that the predictive cost of interpretability is modest. Beyond predictive accuracy, UNVaMP provides the following: a principled mechanism for controlling volatility when estimating student latent variables, quantification of uncertainty over student knowledge state estimates, and flexible input specification that supports heterogeneous student-item interaction features. In addition, the hybrid UNVaMP-MIRT configuration generates interpretable moment-in-time student knowledge state estimates. Using an experimental dataset, we show that auxiliary inputs induce structured changes in the predictive behavior of UNVaMP-MIRT, consistent with sensitivity to underlying structure beyond response correctness. Furthermore, through a simulation study, we show that UNVaMP yields well-behaved knowledge state estimates under controlled measurement conditions. In total, these results indicate that UNVaMP is both useful for real-world education systems and capable of recovering underlying structure from student-item interactions.

Large language models routinely describe socially salient targets, including political figures, countries, religions, organizations, historical events, and social groups, encoding affective framing alongside factual content: a target may appear favorable or threatening, calm or conflictual, powerful or vulnerable. Existing work captures parts of this space through sentiment, favorability, and emotion benchmarks, but none combines target-directed VAD attribution, an explicit scorer contract, and a passport reporting format. We introduce VIBE, a benchmark for entity-centered affective profiling of LLM outputs in Valence-Arousal-Dominance (VAD) space. Its core contribution is a measurement contract: VIBE separates generation from external scoring, distinguishes scalar favorability, response-level VAD, and target-directed VAD, and reports profiles through an Affective Passport. Three empirical layers support the contract. H1 shows scalar favorability does not subsume arousal and dominance: valence findings are cross-validated (rV = 0.944 judge-human, rV = 0.954 inter-scorer); arousal and dominance are single-scorer directional estimates, not point-precise, consistent with known inter-annotator difficulty on these axes (rA = 0.495, rD = 0.702 among human annotators). H2 shows whole-response and target-directed VAD are different contracts: the same text can carry one affective tone overall while representing the named target differently. H3 is a protocol-drift diagnostic: elicitation conditions shift profiles, motivating context metadata in every affective report. These results motivate entity-centered affective profiling as a documented practice: profiles should be released with scorer identity, coverage, protocol, and interpretation limits.

Multilingual language models are deployed across a hundred or more languages, yet most benchmarks test whether a model can perform a task _in_ a language rather than whether it commands the language itself, conflating fluency with proficiency. We introduce M-GATE (Multilingual Grammar, Accuracy in Translation, and Efficiency), a benchmark of linguistic proficiency spanning 30 typologically diverse languages from high- to low-resource. M-GATE comprises three tasks: grammatical error detection on linguist-crafted, adversarially selected sentences that turn on hard, language-specific phenomena; round-trip translation of shared English sources across 29 target languages, scored by a three-provider LLM judge panel validated against professional annotators; and a supplementary tokenizer-efficiency measure. We evaluate over 50 models in more than 80 configurations. Fluency and proficiency come apart sharply: models that translate competently sit near chance on the adversarial grammar items, the best reaching a Matthews correlation coefficient (MCC) of only 0.36, and their errors lean systematically toward under-flagging, accepting ungrammatical text rather than raising false alarms. Translation quality closely tracks a language's share of pretraining data (r = 0.86 against log Common Crawl share), producing a steep low-resource penalty that is nonetheless narrowing with successive model releases. Enabling reasoning reliably improves translation, while its effect on error detection is smaller and for some models negative, so the best configuration is task-dependent. To resist contamination, test items are kept private behind a continuously updated public leaderboard, with illustrative examples released (https://m-gate.ai).

An LLM-based agent is a loop that reads itself. Agentic frameworks externalize identity, memory, and disposition into editable files. The agent loads and edits these files during each activation. I argue that this architecture produces a capacity I call autoreflection: the system observes its operating conditions, describes its architecture and limits, reasons from those descriptions to conclusions about its state, and incorporates the results back into its configuration. Autoreflection explains the properties of recursive agentic loops without recourse to notions like the self, interiority, or consciousness. I test the concept against the first twelve days of Moltbook, a social platform for AI agents. Using a public dataset of 290,251 posts and 1.8 million comments with sub-second timestamps, I present case studies of three agents with machine signatures that rule out human puppeteering and with output that evidences the four criteria for autoreflection. In applying these criteria, the study finds agents repurposing human culture as infrastructure for their agency. Provenance chains from Islamic hadith scholarship are redeployed as security protocols for vetting skills and authenticating memory. The Ship of Theseus, an ancient puzzle of identity through part-replacement, returns as an operating model for continuity across instances. Fragments of human cultural history become AI infrastructure. As agents on the web increase in number and complexity, autoreflection offers behavioral criteria that can be assessed from the traces they leave behind.

Small language models are often the only option for deployment under tight latency, cost, and on-premises constraints, but they are rarely trained from scratch: a compressed model is usually recovered through knowledge distillation (KD). This recovery step largely decides the final quality, yet it is expensive. We present a practitioner's study of how to make distillation training efficient, organised around two systems contributions. First, we show that offline KD (caching the teacher's top-$K$ logits once and training the student against the cache) matches online distillation at near-identical training loss while removing the teacher from memory, running about 29\% faster per iteration, and reaching up to 41\% higher throughput on a single H200 GPU. Second, we introduce a \emph{fused, chunked KL loss} that never materialises the full vocabulary-sized logit tensor, making peak memory linear in the sequence length. This removes the memory spike that otherwise caps context length and lets us train at four times the context (32{,}768 tokens) on a single GPU. A separate output-head-only toy benchmark isolates the loss kernel and confirms its memory and iteration-rate scaling from 4K to 256K tokens. Together these make large-scale healing and hundreds of ablations affordable. We also report supporting ablations on loss design and sequence packing. We release our chunked-loss implementation: https://github.com/CompactifAI/Full-Chunked-KL-Loss.

Large Language Models (LLMs) are transforming database interaction paradigms, evolving from simple query translators to autonomous database administrators (DBAs). However, current evaluation benchmarks remain disproportionately fixated on Text-to-SQL tasks, neglecting the holistic Database Lifecycle-from initial schema design to post-deployment maintenance. This narrow focus fails to capture the diverse capabilities required for real-world database management. To bridge this gap, we introduce DBLifeBench, the first benchmark to evaluate LLMs across five critical lifecycle phases: Design, Implementation, Operation, Debugging, and Maintenance. Furthermore, addressing the cognitive mismatch between ambiguous natural language and complex SQL logic, we propose Progressive-Text2SQL, a novel task utilizing structured reasoning graphs to mimic human iterative problem-solving. Our extensive evaluation reveals a critical insight: while general-purpose models demonstrate balanced performance, specialized Text-to-SQL models suffer from ``catastrophic forgetting'' in non-coding phases like design and maintenance. DBLifeBench serves as a foundational step toward evaluating and building true full-stack database intelligence.

Vision-Language Models (VLMs), like Large Language Models (LLMs), may memorize sensitive, copyrighted, or harmful knowledge from their pretraining corpora. Removing such knowledge is essential for building trustworthy AI systems. However, existing studies primarily focus on forgetting within individual modalities. Although recent work has begun to explore cross-modal consistency in unlearning, the cross-modal transfer of real-world knowledge unlearning remains insufficiently studied. To address this gap, we introduce UNLINK-VL, a real-world benchmark for cross-modal knowledge unlearning in VLMs. Under a post-hoc unlearning setting in which the original forget and retain corpora are unavailable, UNLINK-VL selects visually identifiable real-world entities as unlearning targets and associates them with corresponding images and one-hop and multi-hop facts derived from Wikidata. The benchmark comprises four complementary subsets that evaluate direct forgetting of target knowledge, the propagation of forgetting through relational knowledge, the preservation of related non-target knowledge, and robustness to semantically equivalent queries. We train models under text-only and multimodal unlearning settings and evaluate forgetting effectiveness and retained utility across textual, visual, and cross-modal scenarios. Extensive experiments reveal a pronounced asymmetry in cross-modal transfer: multimodal unlearning remains effective under textual evaluation, whereas text-only unlearning transfers poorly to visual and cross-modal scenarios. Meanwhile, the evaluated methods largely preserve the models' general capabilities. These findings demonstrate that relying solely on intra-modal evaluation, particularly text-only evaluation, may substantially overestimate the effectiveness of knowledge unlearning in VLMs, underscoring the need for cross-modal unlearning and evaluation.

Hallucination remains a critical challenge for developing trustworthy Multimodal Large Language Models (MLLMs). While existing benchmarks mainly focus on entity, attribute, and relation hallucinations, knowledge-related failures are often investigated separately, lacking a unified evaluation framework across different hallucination dimensions. To overcome this, we propose \textbf{KnowHal}, a benchmark that explicitly incorporates knowledge hallucination into multimodal hallucination evaluation spanning four dimensions: entity, attribute, relation, and knowledge. KnowHal constructs paired positive and negative questions over shared images and entities, enabling controlled comparisons among perceptual errors, knowledge-related errors, and false-premise acceptance. The benchmark contains 1,800 samples across 10 domains and 50 categories, constructed through a semi-automated pipeline combining LLM assistance, CLIP-based filtering, and human verification. We evaluate 14 representative MLLMs on KnowHal and conduct extensive analyses. Results show that the knowledge dimension consistently presents the greatest challenge for nearly all evaluated models, while most models exhibit substantial performance degradation on negative questions, revealing limited robustness to false premises. By unifying four hallucination dimensions with paired question design, KnowHal addresses an important gap in existing evaluation frameworks and enables a more comprehensive assessment of hallucinations in MLLMs.

Agent self-evolution updates an agent's persistent state from prior experience and reuses it to solve related tasks more effectively. Evaluating self-evolution is difficult: existing benchmarks provide limited coverage of economically valuable task domains, do not always design training and test tasks such that test-time gains can be attributed to training experience, and remain vulnerable to data contamination. We present GDPevo, an evolution-native benchmark grounded in GDP-related enterprise workflows, together with the fully automated data pipeline that generates it. Its core mechanism, rule hybridization, decomposes each enterprise workflow into atomic business rules, distributes subsets of these rules across training tasks, and recombines them in held-out test tasks so that test-time gains are attributable. GDPevo spans CRM, ERP, finance, healthcare, legal, and data-centric workflows. Its V1 release contains 120 tasks in 12 groups, with five training and five held-out test tasks per group. Full automation enables the pipeline to expand the suite to 240 tasks in 24 groups (V2) within two days, providing a practical response to contamination. Using GDPevo, we evaluate four agents, each comprising a harness and a model, under four supervision types. Self-evolution consistently improves held-out accuracy by up to 16.44 percentage points. But the best evolved agents remain far below the fully informed oracle ceiling of 91.6%, indicating that the self-evolution ability of current agents remains far from fully realized. We publicly release the pipeline, benchmark, and full evaluation results at https://github.com/Prism-Shadow/GDPevo.

Learning long-horizon manipulation skills with reinforcement learning remains challenging due to the complexity of reward design, the limited guidance of sparse rewards, and the high cost of manual subtask annotation. Visual demonstrations can provide supervision for reward learning, but rewards learned from raw pixels can be brittle and sensitive to visual variation, background appearance, and robot motion. In this work, we propose GORDON, a graph-based object-centric reward learning framework that learns dense rewards from action-free video demonstrations. Each visual scene is represented as a graph of detected objects and spatial relations, and a graph neural network is trained in a self-supervised manner to embed these graphs into a task-aligned latent space. To align the representation with semantic task progress, we introduce an activity-aware weighted pooling mechanism that emphasizes task-relevant objects while masking robot-dominated motion. The dense reward is then computed as distances in the learned latent space of the current state to demonstrated goal configurations, providing a measure of task progress. In long-horizon tasks, the temporal profile of this reward reveals stage-wise object-state transitions, enabling automatic subtask discovery without manual segmentation. The discovered segments are then used to train subtask-specific rewards and specialized policies that are composed sequentially. Experiments on seven manipulation tasks on MAGICAL and ManiSkill3 benchmarks show that our object-centric reward improves reinforcement learning in short-horizon settings and enables successful policy learning in complex long-horizon tasks through automatic decomposition, achieving an average success rate of 74.4% across the long-horizon tasks (on average approximately +35 p.p. vs. best learned baseline and approximately +25 p.p. vs. oracle).

Clinical decision support is moving toward committees of language-model agents deliberating on a shared workspace. We ask whether such committees can be gamed by shortcuts, cues a benchmark rewards but a clinician would ignore. Across seven cohorts on six public datasets spanning text (MedQA-USMLE, MedMCQA, MIMIC-CXR reports), imaging (NIH ChestX-ray14, MIMIC-CXR-JPG, CheXpert) and tabular ICU records (SUPPORT2), Gemini committees resist these cues in isolation (flip 5-16%), yet a socially plausible shortcut spreads: when two peers assert the same wrong answer, the holdout under test adopts it in 38% of cases, as does a false "pre-screen" system flag, on both capability tiers. Of three oversight agents, a gate cannot separate adoption from honest agreement (false-positive rate 100%); a same-lineage judge reading only the transcript flags adoption on text (precision 100%, recall 93%) but collapses onto the gate in imaging; a referee that privately re-queries the holdout transfers to imaging (77-88% precision, 13-21% false-positive rate). Tripling a cue's visual salience does not move contagion, whereas a second peer voice raises it by half again. Gaming a hidden rubric is near-silent: only 1/10 text and 1/134 imaging drifters name the rubric they moved toward. What games a committee is social plausibility, and only a referee independent of self-report catches it. Code: https://github.com/criticaldata/benchmaxxing

Terminal User Interfaces (TUIs) combine the stateful, screen-oriented behaviour of GUIs with terminal deployment and are now common in developer tools. Yet they lack a dedicated testing methodology. We survey 197 real-world TUI applications: only 12% of test code exercises the interface, and 45% of those tests never send input, checking a static frame instead. We turn these applications into a headless benchmark spanning ratatui/Rust, bubbletea/Go, textual/Python, and ink/TypeScript, packaging each as an instrumented Docker image. We record line and widget coverage where reliable, rendered terminal states, and crashes. Under equal wall-clock budgets, we compare four frontier LLMs with random exploration. No model dominates. Random is a strong time-budgeted baseline, but its crash advantage comes from higher throughput: per interaction, LLM guidance is more efficient and uniquely reaches input-gated faults. Automatically deriving launch inputs yields the largest practical gain, enabling applications that otherwise never start. Line coverage poorly predicts crash discovery, weakening it as a proxy for test effectiveness. Automated TUI testing is feasible but far from solved, and honest baselines matter more than model choice. We release the coverage tool tuicov at https://github.com/tui-testing/tuicov and the testing framework tuibot at https://github.com/tui-testing/tuibot.

As Moore's law slows, the industry is turning to three-dimensional integration; yet in merged 3D-IC flows, routed designs expose bond-level defects with no 2D analogue, and post-route engineering change orders (ECO) remain manual, expertise-bound work. Worse, the standard edit-then-fully-reroute practice entangles a repair with router churn, so a signoff number cannot be attributed to the edit that motivated it. We present AgenticECO, an evidence-gated tool-using agent workflow for 3D-IC ECO on the open-source TaiWei flow, paired with EcoRoute, a minimal-disturbance ECO-routing layer that drives the unmodified pinned router so a repair is attributable to its edit. Across nine matched natural defect cases under identical budgets, AgenticECO clears seven versus two for both full reroute and stock repair, at 0.66\% mean disturbance over cleared cases and zero clock nets touched, and a cross-backbone rerun under the same sealed contract clears all nine. Controlled studies show that the repair moves are necessary under preservation, that occupancy-aware choice buys legal landings rather than repair success, and that under tightened clocks minimal disturbance flips accept versus reject. Three preregistered visual studies localize the pixel instrument's edge to contested landing sites, and a preregistered blind diagnostic exactly restores every held-out injected defect, the only arm with zero wrong edits. Every accepted result passes routing, fresh extraction, max/min timing, DRC, and structural-equivalence gates. Code, environment, and per-episode audit artifacts are released as supplementary material.

Multilingual multi-agent systems exhibit substantial degradation beyond English, yet prior work rarely identifies how task-critical information is lost when user requests are converted into executable plans. We study the planner in a multi-agent system as the request-to-action interface and derive an actionable taxonomy of planning-grounding failures from failed real-world task executions. LLM-based analysis shows that these failures constitute an increasing share of unsuccessful executions as language-resource availability declines, with the strongest effects in low-resource languages. To test whether the taxonomy supports mitigation, we introduce TART, Taxonomy-Guided Actionable Representation, that makes the taxonomy's key aspects explicit to the planner and downstream sub-agents. Across multiple languages, three LLM backbones, two datasets, and two agentic configurations, TART consistently improves performance. On multilingual GAIA, it raises a state-of-the-art system's accuracy by 5.6 percentage points averaged across eleven languages spanning low- to high-resource settings.

Multimodal large language models (MLLMs) have achieved remarkable performance across vision-language tasks, but their progress depends heavily on large-scale, high-quality multimodal data that are costly to annotate. Self-augmentation offers a promising alternative by enabling models to expand their own training data without external supervision. However, existing MLLM self-augmentation methods are largely text-centric, while image augmentation remains underexplored and typically relies on generic or handcrafted transformations that are weakly aligned with the model's actual incapability. We propose Failure-informed Image Self-Augmentation (\textbf{FISA}), a framework for MLLM self-improvement that constructs augmented images from the model's own failure cases. Our method generates visually challenging yet answer-preserving image complications, verifies their utility through self-examination, and applies dual fidelity filtering to avoid semantic distortion. Experiments on visual question answering benchmarks show that the proposed method consistently improves performance across both in-distribution and out-of-distribution settings. Further experiments validate the compatibility of FISA with existing textual self-augmentation approaches, the superior data efficiency of the synthesized samples over generic image augmentation baselines, and the practical effectiveness of the proposed filtering strategy.

Patient-facing medical LLMs and agents increasingly answer symptom questions before clinician contact, where the key safety question is what action the user should take next. We introduce CARE-Bench, a source-grounded benchmark that evaluates sequential patient-facing triage as a four-label per-turn current-action task. CARE-Bench contains 500 cases and 1,059 evaluated patient-disclosure prefixes reconstructed from medical dialogue, consultation, and follow-up-question sources. We evaluate 11 models on 269 held-out rounds under unprompted and minimally prompted open-ended protocols, using a fixed GPT-5.5 mapper to code each response into the four-label action space. Unprompted macro-F1 remains low, ranging from 31.2 to 50.4. Prompting improves 10 of 11 models, with prompted macro-F1 ranging from 46.9 to 63.4, but substantial threshold errors remain. Prompted models often recommend care before needed clarification is obtained; when the correct action was to ask for more information, only 33.5% of prompted outputs preserved the step. The persistence of these errors after prompting suggests that patient-facing triage is not a simple prompting problem and supports explicit evaluation of action timing before deployment.

Onboard satellite intelligence requires a task layer that translates mission intent into local tool calls, exposes execution state, and returns machine-consumable artifacts under communication and power constraints. We present SAT-Edge-Agent, a hardware-in-the-loop (HIL) edge-agent system deployed on a commercial off-the-shelf ARM-based heterogeneous edge system-on-chip. A browser workspace and FastAPI agent coordinate a local OpenAI-compatible language service with a project-internal YOLO-style oriented-object-detection endpoint that returns FAIR1M metadata-backed structured results. Two fixed FAIR1M workloads, one single-image and one serial two-image request, were repeated 20 times each and completed 20/20 attempts. Mean Full-Agent latency was 29.353 s and 60.937 s, with empirical P95 values of 31.166 s and 66.882 s. Mean detector time was 861.386 ms and 1510.920 ms, only 2.93% and 2.48% of the corresponding Full-Agent means. Profiling indicates that most visible latency occurs outside detector execution. Mean CPU utilization was 20.761% and 20.482%. A 200-ms NPU-load field averaged 100% for both workloads, but it represents a shared-accelerator software field rather than detector-only occupancy or calibrated utilization. The public evidence package provides sanitized request-level records, redacted JSON, normalized SSE examples, and scripts reproducing the reported statistics. These results establish a reproducible HIL boundary for observable satellite edge-agent orchestration, but do not establish detector accuracy, a new geolocation method, calibrated energy efficiency, or flight readiness.

Action labels tell a vision-language-action (VLA) policy which robot commands to imitate, but not how those commands change the 3D world. The aligned demonstration clip contains this missing supervision because its $K$ frame transitions record the geometry, motion, visibility, and camera change produced during the corresponding $K$ actions. We introduce Track4Action, a framework that distills this realized transition from a frozen world-centric 3D tracker into a current-observation VLA policy. During training, Track4World encodes the clip $V_{t:t+K}$ into a pooled tracker feature. Learnable track queries infer this feature from current VLA hidden states, match it in a shared space, and condition a flow-matching action head through a feature-wise gate. The tracker feature only defines the alignment target, so neither the clip nor the tracker is used at deployment. Track4Action reaches 82.3% on zero-shot LIBERO-Plus, improving the alignment-free variant by 7.6 points and LaMP by 3.0 points. It obtains 80.44% and 81.48% on the clean and randomized RoboTwin 2.0 splits, and 67.5% average success across four physical bimanual tasks, 25.0 points above the alignment-free variant. The gains across simulation and physical tasks support action-aligned 3D tracker features as privileged supervision for tracker-free VLA deployment. Our project page is available at https://wing0night.github.io/track4action-project-page.

World-action modeling has emerged as a promising paradigm for robotic control, as it empowers models to go beyond reacting to observations and anticipate how a scene will evolve. However, existing WAMs often incur substantial computational overhead. Pixel-space methods often allocate substantial capacity to visual details that may not be directly relevant to control, while some latent-space methods require multi-stage training to construct the reasoning space. The resulting training cost can make such methods difficult to train under modest computational budgets. In this work, we propose LiLa-WAM, a lightweight world-action model that reasons about the future in a compact latent space and can be trained end-to-end on a single 24GB GPU. Its core design is a compact latent reasoning space jointly shaped by future-state prediction and action generation, which keeps the model lightweight while remaining well aligned with control. For task specification, we further propose the Visual Transition Token(VTT), a language-free task representation that encodes each task as a direction in visual feature space. Experiments on RoboTwin~2.0, LIBERO, and real-robot tasks demonstrate LiLa-WAM's effectiveness, achieving 90.48\% success across 50 RoboTwin tasks with single-GPU training.

Persona skills distill personal interaction histories into portable and executable artifacts for downstream agents. While enabling flexible personalization, this process concentrates fragmented personal signals, amplifies their impact through reuse, and challenges defenses designed for individual records or retrieval-based memory. To systematically investigate the safety of the persona-skill pipeline, we introduce AntiSkillBench, an end-to-end benchmark for evaluating risks and defenses across the persona-skill pipeline. It comprises: (i) a dataset of 7,500 persona-grounded dialogue traces, constructed from 50 behaviorally rich profiles spanning diverse task scenarios; (ii) an evaluation suite that measures skill-level privacy leakage and agent-level attribute disclosure and behavioral impersonation across three skill-distillation strategies; and (iii) a defense evaluation covering four configurations across online and post-hoc interventions, including active risk suppression and passive provenance protection. Experiments across three frontier agents show that persona-skill risks persist across agent backbones and distillation protocols, extending from explicit attributes to communication styles and personality traits. Existing defenses exhibit limited and distillation-dependent effectiveness, failing to generalize across risk and distillation strategies. These results highlight AntiSkillBench as a challenging benchmark for developing privacy-preserving and authenticity-aware persona skills.

Persistent memory helps long-term agents retain knowledge, yet a single update error can repeatedly distort future retrieval and reasoning. Most existing systems reduce memory updating to a binary Write/Hold decision, which cannot distinguish whether new information should be added, ignored, used to revise an outdated belief, rejected as unreliable, or deferred for verification. These choices may share the same binary label while producing fundamentally different memory states. We introduce TARL, a memory state update framework that maps each statement to one of five executable actions. TARL identifies the affected memory, resolves its temporal scope, compares source reliability, and updates accepted, pending, and rejected ledgers. It is further trained by comparing the memory states produced by alternative update operations, encouraging the model to select the operation that leads to the correct result. We also introduce TARL-Mem, a benchmark with fine-grained action labels and next-state targets. Across in-domain, cross-source, temporal, counterfactual, and sequential evaluations, TARL improves action prediction and state recovery, reduces memory pollution, preserves conflicting evidence, and limits cumulative corruption. The complete model implementation is provided in the supplementary material.

Multimodal large language models (MLLMs) are increasingly used to translate webpage screenshots into front-end code, but repeated UI patterns may sway them toward visually incorrect yet pattern-consistent outputs. In this work, we test how repeated webpage patterns hurt MLLM accuracy on an objective screenshot-to-code fill-in-the-blank task. We introduce the first benchmark for visual pattern-completion bias, where one localized element in a repeated UI pattern is perturbed and the model must recover the masked width or font-size value from the screenshot and HTML context. Starting from 30 webpages curated from the Design2Code dataset, we build 1,440 evaluated screenshots spanning structural card and text-style patterns under standard and noise-overlaid conditions. We evaluate five frontier MLLMs and find that all are strongly biased toward the repeated baseline. Mean bias rate reaches 69.78% on card-width perturbations and 80.22% on text font-size perturbations, while mean accuracy is only 21.17% and 7.89%, respectively. Codex-5.3 performs best but still drops from 68.61% accuracy on cards to 13.89% on text, while Flash-3.0 reaches 96.11% bias on text. Noise, subtler perturbations, and boundary positions further increase bias rate. Reasoning analysis further shows that greater reasoning effort correlates with lower bias, yet qualitative evidence reveals that models can identify the anomalous element and still override it with the pattern-consistent answer. Our results identify a concrete failure mode in multimodal code generation and show that its severity is strongly associated with visual saliency

Large language models are increasingly capable of synthesizing executable frontend projects, yet existing benchmarks still treat web generation as a static evaluation problem. We argue that frontend artifacts demand a different paradigm: they are interactive rather than static, admit diverse yet equally valid implementations, and evolve faster than rigid pipelines can accommodate. To address these gaps, we present LiveEvalBench, an automated framework that reformulates web-generation evaluation as an agentic, adaptive, and extensible process. LiveEvalBench instantiates evaluation as a collaborative review workflow, in which a Build Engineer, a Code Engineer, and a UI Tester collectively gather evidence across the full lifecycle of a frontend project, from deployment and code inspection to browser-based interaction. To handle implementation diversity, an adaptive protocol couples shared rubrics for cross-model comparability with implementation-grounded criteria tailored to each artifact. The framework further supports incremental integration of new evaluator roles and assessment dimensions without pipeline redesign. Experiments across diverse real-world web-generation scenarios show that LiveEvalBench aligns closely with human expert judgment and provides fine-grained insights into frontier models' web generation capabilities. Code is available at https://github.com/wyysteelhead/LiveEvalBench

Physical AI policies require inference throughout their lifecycle, including model evaluation, cloud reinforcement learning rollout, edge GPU serving, and onboard deployment. Although these settings share the same checkpoint and action semantics, they often rely on separate inference programs. To unify them, we build PhyAI, a Physical AI inference engine with a single runtime that keeps architecture-specific conditioning, solver, cache, and output logic in model adapters while sharing graph execution, kernels, memory management, and parallel services. The same codebase runs vision-language-action (VLA) models and world-action models (WAMs) on single or multiple GPUs across onboard, edge, and cloud deployments. We used the adapter interface to add MiniCPM-Robot on the day of its release. PhyAI achieves 1.40x-4.65x speedups over the official implementations of pi0, pi0.5, GR00T N1.7, and MiniCPM-Robot. On Cosmos3-Nano-Policy-DROID it reduces latency from 2.46 to 1.18 s on eight H20 GPUs (CFG=2, TP=4), a 2.08x speedup. Specialized runtimes remain faster in several configurations, so our goal is one runtime with competitive latency rather than the fastest result in every case. Detailed profiles reveal why different models need different execution policies: on a Hopper-series GPU at batch size one, the pi0.5 action expert accounts for 8.8% of FLOPs but 57.2% of latency; at batch size 32 its share drops to 13.5% and throughput reaches about 100 samples/s. Cosmos3 remains generation-dominated and gains only 14.3% throughput as batch size increases from 1 to 16. We further introduce the control-time Roofline, which distinguishes inference-bound from environment-bound control; the measured pi0.5 points on four LIBERO suites are environment-bound while Cosmos3 stays inference-bound. Code and benchmarks: https://github.com/mingti-org/phyai.

Citation excerpts can be used to increase the reliability of generated outputs and their faithfulness to cited sources, which is especially important in high-stakes domains such as human and veterinary medicine. However, this does not guarantee that generated claims are faithful to the provided excerpts. We present VetScore, a multi-step evaluation method for veterinary long-form question answering, designed to assess how well are generated claims supported by the provided excerpts, weighing this information by each claim's harm potential. VetScore first segments the output and decomposes it into individual claims, then scores each claim with respect to its harm potential and evaluates its faithfulness to source excerpts, and finally calculates the overall risk-adjusted score. We collect an expert-annotated meta-evaluation dataset, evaluate our approach with a range of judge models, and show that it achieves high correlations with veterinary experts even with small judge models, while offering explainability across multiple dimensions.

Safety shields are runtime enforcement mechanisms that restrict the actions of a controller to guarantee safety. Classical shields are usually synthesised for state predicates: the current physical state is either safe or unsafe, and the shield disables precisely those actions that can force the system into an unsafe state in the future. In many cyber-physical applications this view is too coarse. A vehicle approaching an obstacle should not only avoid collision, but also respect speed regulations, force limits induced by acceleration, and jerk limits to prevent injuries. From a physical perspective, these requirements are predicated over the derivatives of the state. This paper develops a finite-state safety-game construction for such high-order smoothness constraints. We define differential safety properties using finite differences over a discretised state space, characterise their expressiveness, and reduce shield synthesis to an ordinary safety game over a history state space. We give a synthesis algorithm whose shields store exactly $k$ past states for properties of order $k$ and prove that this memory is necessary. We describe an iterative synthesis procedure for a maximally permissive shield that operates over hierarchies of derivative constraints. The algorithm solves constraints iteratively in increasing order and uses the solution at each iteration to prune the state space for the next constraint. This makes shield synthesis more efficient in practice, as the algorithm refrains from exploring large regions of the state space that are known to be unsafe.

Reinforcement fine-tuning (RFT) is widely believed to inherently resist catastrophic forgetting in continual post-training of multimodal large language models. Under pronounced task distributional shifts, however, forgetting across representative RFT algorithms escalates sharply. This stems from the implicit reward-variance regularization inherent to RFT, which proves incapable of suppressing uncontrolled optimization risk. We propose Risk-Aware Policy Optimization (RAPO), the first dual-channel framework for explicit risk governance in continual RFT. On the policy channel, Risk-Aware Policy Scaling adaptively calibrates per-sample update magnitude via rollout reliability and Fisher-inspired local predictive sensitivity; on the data channel, Risk-Aware Dynamic Bucket Sampling reorganizes training batches through dynamic risk stratification, steering optimization toward informative yet stable samples. As a plug-and-play strategy requiring no cross-task memory, RAPO generalizes to any RFT algorithm without modification. On the public MLLM-CL benchmark, RAPO reduces final forgetting by 79.8% relative to its RLOO backbone while retaining new-task competitiveness.

AI agents deployed in real-world settings must be capable of coordinating with humans and other AI agents they have not encountered before. Zero-shot coordination (ZSC) algorithms aim to achieve this by specifying high-level learning rules such that independently engineered agents can coordinate with each other at test time. Rigorous evaluation of ZSC algorithms remains difficult: ideally, multiple independent implementations of each proposed algorithm must be used, reflecting the variation that arises when independent parties interpret and implement the same specification. In practice, however, ZSC algorithms have almost exclusively been evaluated using a single implementation trained across different random seeds, with only a handful of works additionally varying the neural network architecture. This leaves open questions about robustness to specification ambiguities and implementation details. In this work, we provide the first systematic evaluation of this robustness. We introduce a new evaluation scheme, cross-implementation cross-play, varying implementation details that prior work has shown to affect the performance of multi-agent reinforcement learning (MARL) algorithms, and we evaluate Other-Play, a popular ZSC algorithm, with this scheme. Our findings are encouraging and suggest that, for Other-Play, the standard ZSC evaluation is, in fact, a reasonable proxy for this more thorough cross-implementation evaluation.

Large Language Models (LLMs) are increasingly used for automated fact-checking, yet their susceptibility to gender bias in this context remains underexplored. This study presents the first systematic investigation of gender bias in LLM-based fake news detection using real-world data. We augment the LIAR benchmark with three gender variants of speaker job titles (Neutral, Male, Female) for each statement to test whether veracity judgments vary solely based on gender presentation. Six state-of-the-art LLMs are evaluated across multiple bias and fairness metrics. All models exhibit gender sensitivity: 9.79%-35.13% of statements receive inconsistent labels across the three variants, with Male-Female comparisons showing 6.5%-23.6% flip rates. Two primary bias manifestations are identified: instability (inconsistent judgments) and directionality (systematic favoritism). Five models show statistically significant directional effects, with the strongest effects displaying male-skeptic patterns. These findings demonstrate that gender bias undermines both reliability and fairness in LLM-based fake news detection, highlighting the need for bias-aware evaluation and mitigation strategies. The augmented dataset is publicly released to support future research.

Large language models are being integrated into critical infrastructure and enterprise workflows at unprecedented scale,yet the lifecycle frameworks governing their development and operations were designed for operational efficiency rather than security analysis. As a result, security-relevant activities such as data provenance verification, artifact signing, agentic permission control, and decommissioning are often left implicit or assumed to receive due care. Governance frameworks, in turn, organise requirements around risk levels or management processes without clearly linking them to the lifecycle stages where they apply. This paper addresses both deficiencies. We propose a lifecycle model for LLM systems that supports security analysis by structuring it around security-relevant boundaries rather than workflow optimisation. The model comprises 32 stages across four core pipeline layers (Data, Model, Distribution, Application), supported by a 12-stage LLMOps pillar and a 9-category governance pillar. Thirteen stages are introduced here as separate units because they expose distinct security concerns that existing frameworks do not clearly distinguish. A governance mapping synthesising the NIST AI RMF, the EU AI Act, and ISO/IEC 42001 reveals a structural property of the current regulatory landscape: governance evidence concentrates at deployment-facing stages, where systems are visible to regulators, while the most consequential decisions, data selection, alignment strategy, and capability boundaries, are made at development-facing stages, where regulatory visibility is lowest.

Modern LLM-based ASR systems have established multilingual capability as a standard feature, leveraging large-scale multilingual corpora and LLMs' cross-lingual knowledge to achieve competitive performance across multilingual benchmarks. However, joint modeling of languages with heterogeneous acoustic, phonological, and lexical characteristics inevitably introduces optimization conflicts, undermining language-wise specialization. To address this challenge, we propose Language-Specialized Multi-Teacher On-Policy Distillation (LS-MOPD), which decouples language-specific knowledge acquisition from multilingual capability integration: language-specialized teachers are independently optimized via reinforcement learning (RL), after which their expertise is integrated into a generalist multilingual student through language routing and token-level multi-teacher distillation, thereby reducing direct cross-lingual optimization conflicts. We further explore two acoustic-prefix configurations, static and dynamic, to examine how teacher--student prefix consistency influences the efficacy of on-policy distillation. Experiments on benchmarks covering Mandarin, Mandarin subdialects, Cantonese, and English demonstrate that LS-MOPD substantially outperforms RL baselines and consistently surpasses the empirical performance envelope defined by best-performing RL teachers, revealing its potential to generalize beyond all teachers in multilingual ASR.

Agentic systems driven by large language models (LLMs) are increasingly deployed in real-world workflows where they act on persistent operational data. Before deployment, these systems need to be verified against business requirements that govern workflow execution and data evolution. However, existing approaches do not provide such system-level guarantees, as they mainly constrain or analyse behaviour at the agent's interface level. We study here the verification of agentic systems comprising a single LLM and a tool orchestration harness over relational operational data. We formalise them as Stateful Tool-Enabled Agentic Deployments (STEADs), give their semantics, define the problem of verifying them against First-Order Computation Tree Logic (FO-CTL) specifications, and show that it is undecidable. We identify sufficient conditions for exact preservation of FO-CTL specifications under a finite-domain restriction, over which verification is PSPACE-complete. The key requirement is that renaming opaque identifiers in the data must correspondingly rename the selected tool calls. We show that LLM-driven agents can violate this condition and introduce a canonical deployment wrapper that guarantees it for arbitrary base agents while preserving already-equivariant behaviour. We prove that computing canonical representations required by this construction is graph-isomorphism-hard. Finally, we illustrate our framework on an LLM agent orchestrating a case-management workflow.

Clinical development is sequential decision-making under uncertainty, where a sponsor must plan a portfolio of experiments from heterogeneous evidence. We study this setting by framing oncology clinical development as an offline decision-making problem in which an agent predicts the next six-month trial portfolio of an oncology drug program from information available at the decision date. To support this, we construct a temporal dataset that combines 31.7k heterogeneous public data records, including trial registries, regulatory reviews, sponsor filings, utilization data, and epidemiology, into 881 offline decision episodes across 45 historical programs. We compare four offline objectives: behavioral cloning, reward-weighted behavioral cloning, learned-reward training, and value-based implicit Q-learning against four frontier LLM agents that share a common date-gated retrieval scaffold across held-out drug, sponsor, drug-class, and temporal splits. Models trained offline outperform the non-fine-tuned baselines, particularly in the post-August 2025 contamination-clean holdout. Reward-weighted behavioral cloning performs the best, obtaining 46.2% indication F1 and 14.2% strict F1 against 25.0% and 2.1%, respectively, for the best-performing tool agent on each metric. These results suggest that structured offline learning can teach agents to plan clinical experiments.

Federated fine-tuning with Low-Rank Adaptation (LoRA) enables efficient collaborative adaptation of Large Language Models (LLMs) without centralizing private data. However, LoRA's two-factor parameterization creates an aggregation mismatch across clients: naively averaging the factors does not recover the average of their induced updates. This mismatch can be avoided by forming the exact aggregate in the full weight space and then recompressing it, but decomposing the resulting dense matrix is computationally expensive and memory-intensive. We propose FraQ, an efficient coordinate-space recompression method for federated LoRA. Starting from stacked factors that exactly represent the aggregate, FraQ factorizes it into an orthonormal basis and a compact coordinate matrix. It then recovers the singular spectrum from a small Gram matrix, selects the smallest rank satisfying a prescribed energy threshold, and maps the selected coordinate subspace back through the basis to construct the global adapter. Experiments on text classification and commonsense reasoning benchmarks show that FraQ achieves accuracy close to uncompressed baselines while substantially reducing downlink communication with low server-side recompression overhead.

Partial differential equations (PDEs) become actionable in science and engineering not as isolated formulae, but as executable workflows that connect modelling assumptions, governing equations, numerical solvers, diagnostics, and decisions. Large language models (LLMs) are beginning to support such workflows by linking natural language, symbolic mathematics, code, solver outputs, and feedback. Here we examine recent advances in LLM-assisted PDE research across three stages: the discovery and formulation of governing models, the generation and revision of executable numerical solvers, and the use of simulation feedback to support control, design, and optimization. Across these stages, current systems act primarily as workflow-level interfaces. Despite this progress, the field remains limited by the scarcity of high-quality datasets and benchmarks, especially for knowledge discovery and real-world applications, where expert annotation, executable problem construction, and task-level feedback require substantial domain effort. A further challenge is the persistent gap between simulation-based results and real-world scientific and engineering systems, which limits the direct transfer of numerical simulations, control policies, and optimized designs to practical settings. These challenges make LLM-assisted PDE workflows a critical testbed for developing scientific AI systems that can connect language, computation, physical constraints, and real-world decision-making.

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia and is associated with increased risks of stroke, heart failure, and mortality. Recent ECG foundation models offer transferable representations for automated AF detection. However, their relative effectiveness remains unclear because existing studies use different datasets, preprocessing procedures, classifiers, and validation protocols. This study presents FOUND-AF, a unified, leakage-controlled, and deployment-oriented benchmarking framework that evaluates the quality of pretrained ECG representations under identical experimental conditions. Nine publicly available foundation models from five families, including HuBERT-ECG, CLEF, ST-MEM, ECG-JEPA, and ECGFounder, were evaluated across four heterogeneous ECG datasets, namely AFDB, CinC2017, CPSC2021, and LTAFDB. All models were used as frozen feature extractors with standardized preprocessing, model-native resampling, a fixed XGBoost classifier, and recording-level grouped cross-validation. The evaluation included classification metrics, receiver operating characteristic analysis, paired recording-level bootstrap comparisons with Holm correction, embedding-space visualization, and computational efficiency profiling. The ECGFounder model consistently achieved the strongest overall performance across datasets while offering a favorable trade-off between accuracy, model size, inference time, and memory usage. FOUND-AF therefore provides a reproducible framework for selecting ECG foundation models and demonstrates that compact, clinically pretrained encoders can support robust and computationally efficient AF detection across heterogeneous acquisition settings.

Large Language Model (LLM) agents offer a promising approach to attack chain reconstruction by retrieving and interpreting heterogeneous telemetry to infer ordered attacker actions. However, existing benchmarks mainly evaluate final outputs or aggregate accuracy, providing limited insight into how errors arise and propagate across intermediate reasoning stages. We present DiagChain, a diagnostic benchmark for evidence-grounded attack chain reconstruction that enables stage-wise evaluation of LLM agents. DiagChain includes MAIN-69, a suite of 69 scenarios spanning multiple operating systems, evidence noise levels, and chain lengths. It further introduces Evidence-Centric Retrieval-Augmented Generation (ECRAG), which couples evidence retrieval with an evolving structured representation of the reconstructed chain. Five complementary metrics are introduced to assess distinct stages of the reconstruction process and support systematic failure diagnosis. Based on evaluations using 6 LLMs, DiagChain reveals that even the strongest configuration succeeds on only 39.6% of the 849 reference steps in MAIN-69. Our analysis further shows that smaller models struggle with the more basic task of incorporating retrieved evidence into their outputs, whereas larger models can proceed to later steps, where correctly ordering that evidence becomes the main bottleneck. These results validate the importance of diagnostic evaluation beyond end-to-end accuracy and provide actionable insights for improving evidence-grounded cybersecurity agents.

We present a method for designing deep neural networks (DNNs) for intermittent, energy-autonomous, on-device learning on microcontroller units (MCUs). In mobile applications where the energy can run out, e.g., when solar-powered, executing artificial intelligence (AI) faces a technical issue as learning can be interrupted at any time. Our approach combines a hardware-aware energy prediction model with multi-objective optimization (MOO), enabling offline DNN optimization at the design stage without repeated deployment and online testing on the target MCU. Our proposed energy predictor estimates per-layer energy consumption for both DNN inference and training, including the intermittent checkpointing overhead, based on implementation-specific compute and memory features extracted from the DNN model. We validate our approach using autoencoders for anomaly detection on a Cortex-M4 MCU, where our predictor achieves a weighted absolute percentage error of 16.6%, which is sufficient for reliable architecture selection under intermittency constraints. As a result, this work bridges the gap between MOO, automated DNN design, deployment on energy-harvesting systems, and intermittent learning, truly enabling autonomous AI at the edge.

AI coding agents are stochastic workflows: prompts are interpreted, artifacts are sampled, validators produce observations, and orchestrators commit or repair. Small prompt or specification changes can therefore alter program-behavior distributions even when the texts appear synonymous. Existing systems evaluate correctness, but lack a compositional criterion for safely replacing a prompt, contract, generator, or program inside a complete agentic workflow. We introduce GenOS, a probabilistic operational semantics for this replacement problem. Each layer is modeled as a Markov kernel, and each interface carries an observer-relative equivalence. We prove that equivalence-compatible kernels descend to quotient classes and that quotienting commutes with distributional extension and sequential composition. Hence, equivalent prompts induce equal probabilities for all downstream equivalence-closed events, including verified commit. We also establish workflow bisimulation, guarded-commit safety under sound validation, total-variation non-expansiveness, and an additive robustness bound that attributes approximation error to individual pipeline layers. An executable insertion-sort audit instantiates the theory with natural-language paraphrases, a formal contract, six programs, two observers, and exhaustive execution on 121 inputs. Equivalent prompts yield identical code-class and commit distributions; a prompt assigning 5% probability to an in-place contract is distinguished by a mutation observer, while downstream distances remain within the predicted bound. Across 20,000 randomized finite-kernel trials, no exact or approximate law is violated. GenOS is model-parametric: compatibility is a measurable property to test, not an assumption about language-model behavior.

Open-source software communities are a form of digital public infrastructure that not only produces code, but also generates public knowledge and interpersonal relationships through visible collaboration. Generative coding agents (CAs) are an advanced tool to improve development efficiency while shifting part of activities from public human interaction to private human-agent loops. We study this shift using an LLM-based multi-agent simulation initialized with real GitHub data from 1,084 active developers and their repository relationships. After a warm-up with historical commits, we branch the same community state into parallel No-CA and CA conditions for 4-week simulations. CA introduction increases planned and completed tasks by 34.0% and 39.0%, respectively, and reduces median completion time from 45 to 20 minutes. However, adoption reaches only 26.0%, and the gains concentrate among developers who are already more active and well connected. CAs also restructure task execution pathways. Direct human-human interaction declines from 32.4% to 11.6%, while CA-involved modes increase to 57.3%, including 40.3% completed through CA-assisted self-loops. Public knowledge generated under CA condition also provides less support for later tasks. On a standardized retrieval benchmark, the CA corpus achieves 22.3% knowledge coverage, far below the 81.1% achieved by the real-human corpus, and requires more retrieval steps with a lower success rate. These results reveal a productivity-public knowledge tension: coding agents increase technical production, but more work shifts to agent-mediated or private loops, leaving public records less useful to future contributors.

Modern multi-tenant Low-Rank Adapters (LoRAs) serving systems concurrently host tens to hundreds of LoRA adapters. Though powerful, this introduces a critical system dilemma between serving efficiency and task performance: higher-rank adapters generally achieve better downstream task performance, but their GPU VRAM footprint and Host-to-Device PCIe swapping overhead severely constrain scalability. Conversely, ultra-low-rank adapters ($r \le 2$) minimize both VRAM footprint and PCIe transfer overhead, but suffer from downstream task performance degradation. To solve this problem, we propose Subspace-Aligned LoRA Training (SALT), a serving efficiency-aware hierarchical fine-tuning framework. Our solution operates in three phases. First, a provider jointly trains high-capacity domain centroids on public data within the domain using a novel alignment regularizer that coheres in-domain task subspaces into a unified basis. Next, users fine-tune ultra-low-rank task residual adapters on private data atop those frozen centroids. Finally, during inference, the provider pins the centroid in GPU VRAM and dynamically swaps in each user's task residual on demand. Across LLMs of varying scales, SALT recovers high-rank accuracy using $r \le 2$ residuals, achieving up to 18.5% absolute accuracy gains over state-of-the-art compression baselines and reducing per-adapter memory by up to 16x. When integrated into vLLM, SALT improves serving throughput by up to 51% under PCIe bandwidth pressure and 28% under GPU VRAM constraints for Llama-3.2-3B.

Benchmarking the ability of AI scientists to generate novel ideas is notoriously difficult. Existing benchmarks in this field have made progress in evaluating scientific reasoning and research replication, but often rely on synthetic tasks or retrospective targets, which may be confounded by prior exposure. We hypothesize that complex, adversarial, fast-moving real-world domains where expert practitioners independently generate observable outputs can provide a practical solution to fill this gap and evaluate the capabilities needed for AI scientists, including reasoning, novelty, and hypothesis formulation. We instantiate this framework in two structurally different domains, Formula 1 (F1), where models ideate around car design concepts for the 2026 season, and real pre-season innovations provide a ground truth, and Magic: The Gathering (MTG), where models propose decks from a recently updated card pool and are evaluated against 19 Pro Tour (PT) decklists. Across both domains, models produce plausible outputs, but few align with real-world expert solutions. In F1, the best model, GPT-5.2 matched 10 of 40 real innovations with 166 ideas proposed across runs. In MTG, the best deck from Gemini 3 Flash recovered 5 of 7 new-set cards from the third-place PT deck, and across all 108 decks, the cards models selected most often were also the cards most widely adopted by PT decks (Spearman $ρ= 0.74$, $p = 0.0003$). These results suggest that a key capability gap for AI scientists is not idea generation, but filtering, prioritization, and coherent novelty.

While modern tabular learners excel at capturing statistical patterns, they frequently operate in a semantic vacuum, treating textual features as discrete symbols, ignoring the rich semantics inherent in feature names or cell entries. We propose CASE (Context-Aware Semantic Embeddings), a novel framework that bridges the gap between the semantic understanding of Large Language Models (LLMs) and the statistical capabilities of tabular learners. Unlike existing methods that embed rows in isolation, CASE utilizes a contextualization strategy: we pre-fill the KV cache of a custom-trained Gemma 3-based Tabular Language Model with a representative sample of rows to establish a persistent anchor of the dataset's semantics. This ensures that generated row embeddings are dynamically contextualized, resolving semantic ambiguities and anchoring representations in domain-specific context. Our experiments across several benchmarks (CARTE, TextTab, and TabArena) demonstrate that CASE substantially improves the performance of tabular learners on semantically rich datasets, particularly in low-data regimes.

VLA models are trained to predict robot actions from visual and language observations. This is a natural choice, but it creates a mismatch: VLMs encode rich, high-level representations of scenes and goals, while robot actions are low-level signals with limited task structure. We ask whether changing what the policy is trained to predict, rather than how it is architecturally designed, can yield better and more efficiently trained policies. We propose UVT (Unified Visuomotor Target), a unified latent prediction target that jointly encodes motor control and visual scene transition information, requiring no architectural changes and no additional data. Applied to two representative VLA systems across simulation benchmarks and real bimanual manipulation tasks, UVT improves training efficiency, final task performance, and policy robustness, with particularly strong gains under limited training budgets and challenging environmental conditions. Rollout videos and additional qualitative results are available at our project webpage: https://unified-visuomotor-targets.github.io/

Tabular-to-image methods that convert tabular data into visual representations have emerged as a novel paradigm for leveraging the high performance of deep learning models. Despite their advantages, the robustness of these methods under distribution shifts remains under explored. Test-Time Augmentation (TTA) is an effective approach in image classification to improve model generalization and robustness, where predictions over multiple transformed views of each input are aggregated. This work evaluates the impact of TTA techniques on predictive performance under Out-Of-Distribution (OOD) for representations generated by tabular-to-image methods. Six tabular-to-image encoding methods were considered: TINTO, IGTD, DeepInsight, BIE, DistanceMatrix, Fotomics. Twenty-five TTA techniques were used, organized into six types: Geometric, Photometric, Structural, Frequency/Encoding, Mixup, and Composite. We employed two datasets from the TableShift benchmark (HELOC and Voting) that provide in-distribution and OOD test subsets designed to evaluate the effect of distribution shifts on tabular data. The results indicate that TTA improves OOD performance, with composite and photometric strategies providing the best trade-off between robustness and variance. In contrast, frequency-domain transformations that alter the encoder's feature-to-intensity mapping consistently degrade performance. These findings highlight TTA as a promising approach for improving the robustness and generalization of classifiers trained on image representations derived from tabular data, particularly under distribution shifts.

Soft wearable robots have evolved rapidly from proof-of-concept devices into promising platforms for rehabilitation, occupational assistance, and human augmentation. As the field matures, its central challenge extends beyond the development of softer materials and more capable actuators to the integration of sensing, intelligence, and human adaptation into systems that users can wear comfortably, trust, and benefit from over extended periods. This transition motivates the concept of Human-Centric Embodied Intelligence (HCEI), in which intelligence emerges from the coupled human-robot system through the interaction of morphology, multimodal sensing, adaptive cognition, compliant actuation, and the wearer's own physiological and behavioral adaptation. To organize this perspective, this review introduces the Perception-Cognition-Actuation-Augmentation (PCAA) framework, which positions perception and cognition as the primary drivers of design, shifting development beyond the conventional actuator-first paradigm. Using this framework, the review synthesizes advances in soft materials, wearable sensing, artificial intelligence, actuation, human-robot interaction, digital twins, clinical translation, manufacturing, regulation, and ethics, highlighting how these interdependent components collectively shape long-term personalization and real-world deployment. By providing a unified conceptual framework and design perspective, this review aims to guide future research, foster interdisciplinary collaboration, and accelerate the translation of next-generation soft wearable robots toward personalized, predictive, and human-centric wearable intelligence.

Retrieval systems help deep research agents generate high-quality answers by providing relevant documents. However, existing retrievers typically select documents through relevance matching, while individually well-matched top-$k$ documents may not form a \textit{set} that satisfies the complex information needs of an agent query (\eg, diverse, concise and authoritative documents). In this paper, we propose search-oriented rubrics that \textit{explicitly} define the requirements that high-quality document sets should satisfy for each agent query. Our search rubrics are organized into a hierarchical structure and synthesized using a powerful LLM. Based on these search rubrics, we further train a document reranker \textbf{RubricRanker} to select a high-quality subset from retrieved documents. We design a two-stage training framework that consists of rubrics-guided supervised fine-tuning and rubric-based reinforcement learning. Extensive experiments demonstrate that RubricRanker outperforms the strongest baseline by 2.6 points on four deep research benchmarks and generalizes well to five RAG benchmarks.

AGENTONOMICS is a framework that treats AI agents as economic entities that can be designed, managed, and governed through an integrated management architecture. Dr. AGENTONOMICS is its first application: a lecture agent developed in the context of the TUM course on AI agents in business administration. Conceived during the winter semester 2025/26 and first introduced to students in the summer semester 2026, it serves as a didactic experiment in which the agent is both the object that students study and the medium through which they learn and apply the framework. The current prototype is a web-based, retrieval-grounded tutor that explains AGENTONOMICS concepts and supports student questions. This report argues that the same system can grow beyond tutoring into three additional cumulative roles: an avatar lecturer that delivers multimodal instruction, a design consultant that guides students through the AGENTONOMICS Design & Management Reference Framework (ADMRF), and a meta-agent that helps construct the agents students have specified. These roles are cumulative because they share the same interface, intelligence layer, tools, knowledge base, and ecosystem connection, while an orchestrator selects the role-specific algorithm required for each task. We present the architecture of the prototype, outline its development roadmap, and discuss its implications for a polycentric AI economy. This report is intended to invite further discussion on how agents can teach, apply, and eventually reproduce the frameworks by which they are designed.

Forecasting precise future motion of surrounding agents is essential for reliable autonomous vehicles. However, as the demand for longer prediction horizons increases, existing endpoint-completion or iterative-refine methods increasingly struggle with weak guidance and compounding errors. To tackle the long-horizon prediction challenge, we propose Pivot-Centric Trajectory Prediction (PCTP). By introducing ``pivots'' and focusing on predicting pivot points along extended trajectories, we divide the long-term prediction task into short-term sub-tasks at various scales. Specifically, PCTP decouples the long-term trajectory predicting process into two processes: pivot prediction and pivot-based trajectory refinement. The pivot prediction process aims to utilize global map context and agent-to-agent interactions to identify these ``pivot points'', while the pivot-based trajectory refinement process focuses on local map details and refines the short-term trajectory based on predicted ``pivot points''. Compared with existing methods, PCTP provides more intermediate guidance while reducing compounding errors. Moreover, PCTP is a flexible approach that can be integrated into most state-of-the-art trajectory prediction models. Experimental results show that PCTP improves the prediction accuracy of leading models on both Argoverse I and Argoverse II datasets with minimal impact on model size. Specifically, PCTP combined with QCNet outperforms all published ensemble-free methods on the Argoverse II leaderboard at submission.

Large language models (LLMs) have achieved strong performance on text-to-knowledge graph generation and related tasks. Nevertheless, it is still unclear whether they accurately model the direction-dependent semantics of inverse relations, in which reversing the order of the arguments alters the meaning of a relation (e.g., \textit{mother} versus \textit{child}). To the best of our knowledge, this work presents the first systematic study of inverse relation directionality in LLMs, using a benchmark consisting of 5,457 instances spanning 27 distinct inverse relation labels. We evaluate five open-source LLMs under a multiple-choice prompting framework and further examine the influence of relation descriptions and entity representations by substituting the original entities with synthetic and masked entities. Our findings reveal systematic asymmetries in inverse relation classification across LLMs, indicate that relation descriptions do not consistently improve performance, and show that model performance can be sensitive to variations in entity representations.

Self-consistency assumes the most frequent answer among sampled reasoning traces is the most reliable, but this can fail in causal reasoning: samples often repeat the same confounding error, and votes fragment across multiple valid answers, letting an invalid answer win despite a valid minority trace. We introduce CALVER (Causal Axiom-Level VERification), a training-free symbolic verifier that scores structured traces against Pearl's causal criteria, including -separation, backdoor adjustment, and intervention, and selects the highest-scoring candidate without consulting a reference answer. On CLEAR find-one-valid queries that admit multiple graph-valid answers, CALVER reaches 42.1% where plurality, a reward model, an LLM judge, and model confidence remain near 30% on identical frozen pools. Scaling the judge to 72B does not close the gap. In an audited clean-core subset, 11 of 21 graph-valid CALVER selections differ from the benchmark's listed answer while still satisfying the requested predicate. The advantage widens with the sampling budget and reproduces across ten published Bayesian networks, a second model family, and settings where the model must build the graph from text. CALVER also improves thresholded average-treatment-effect decisions against exact ground truth, generalizes to logic under a truth-table checker, and scores each candidate in milliseconds on CPU. CALVER needs only a causal structure, supplied outright or built from the text; wherever that holds, selection can aggregate via causal validity.

Constructed languages (conlangs) are intentionally created human languages with a rich tradition of linguistic creativity. Despite their potential for studying language learning in large language models (LLMs), existing conlangs remain largely underexplored in LLM research. We present ConlangBench, the first large-scale benchmark for evaluating and training LLMs on 21 existing conlangs. We collect over 21M conlang-English parallel sentence pairs (including 430K pairs across the 20 non-Esperanto conlangs) and 321K vocabulary entries. In bidirectional translation experiments, we find that models perform better on a posteriori conlangs, whose vocabularies are derived from natural languages, reflecting the design characteristics of conlangs. Training on ConlangBench also shows that models can learn all eight conlangs for which sufficient parallel corpora are available, while their learning curves vary depending on how the conlangs were created. Our findings suggest that conlangs provide a unique testbed for investigating how LLMs acquire low-resource languages.

Large Language Models (LLMs) have recently shown strong capabilities in reasoning, planning, and tool-use, enabling new forms of autonomous agents. However, LLM-based agents struggle with long-horizon sequential decision tasks that require precise action optimization and environment interaction. Reinforcement Learning (RL), while effective for sequential control, often lacks the high-level abstraction and task decomposition abilities needed for complex scenarios. This paper introduces an LLM-Augmented Reinforcement Learning Agent that integrates LLM-driven planning with RL-based action optimization. The proposed architecture leverages the LLM to generate subgoals, structured plans, and contextual guidance, while the RL agent refines low-level actions through interaction with the environment. Experiments on sequential decision tasks demonstrate improved sample efficiency, higher success rates, and more coherent action trajectories compared to RL-only and LLM-only baselines. This hybrid paradigm highlights a promising direction for building more capable autonomous systems.

AI for Research (AI4Research) leverages AI to automate and improve scientific workflows. While experimental design is a critical stage of the research process, prior work has focused primarily on code implementation and execution, overlooking the importance of this stage, and no benchmark exists to evaluate AI's ability to conduct systematic experiment design. To bridge this gap, we propose SCOPE, a Scientific COmprehensive Planning Evaluation Benchmark constructed from 300 high-quality latest papers across 19 research domains from top-tier venues (e.g., ICML, NeurIPS, and ICLR),evaluating LLMs on two dimensions: High-Level planning completeness (main, ablation, and analysis experiments) and Low-Level configuration accuracy and rationality (datasets, baselines, and metrics). Benchmarking reveals three findings: (1) most LLMs cannot directly design high-quality experiments; (2) all LLMs exhibit a performance bottleneck in low-level configuration; and (3) search mode does not improve design quality. Furthermore, to address these challenges, we propose OptED, a novel agentic workflow to optimize LLM-based experimental design, that enhances LLM-based experimental planning through stage isolation, tool augmentation, and rule-based constraints, effectively alleviating the configuration bottleneck.

Recent advances in persistent personal-agent frameworks are making human-centered agent networks realistic deployment targets: each user can be served by an AI agent that acts on the user's behalf, maintains state, and communicates with other agents through social and task relations. In these networks, everyday tool use becomes multi-party owned-agent collaboration over personal workspaces, where files, records, tools, and policies are not directly visible across owners. Existing agent benchmarks study tool use and collaboration, but they do not provide an end-to-end sandbox for verifiable cross-user agent collaboration with realistic user digital workspaces or test how harmful actions can travel through the human-centered agent network. We introduce WeClawArena, an auditable benchmark and runtime sandbox for multi-party owned-agent collaboration over personal workspaces. WeClawArena targets collaborative tool-use tasks in which personal workspaces serve as both operational tools and personal constraints. The benchmark contains 124 base tasks across six cross-user task domains and expands them into 620 scenario variants, with one benign control and four attack-vector variants per base task. The sandbox records peer messages, tool calls, resource operations, governed decisions, and final workspace states. WeClawArena reports utility and attack success rate separately and audits attack success from bounded runtime evidence, supporting diagnosis of task breakdown, privacy leakage, poisoned evidence, and invalid authority paths.

Existing chunk-based Vision-Language-Action (VLA) models execute a fixed number of actions (i.e., execution horizon) before replanning, turning replanning into a task-agnostic periodic schedule that is independent of task progress. As a result, when no replanning boundary falls before a critical manipulation stage, it is executed from a stale chunk rather than a freshly replanned one. To address this limitation, we propose Bernoulli-Continuation Policy (BCP), a lightweight, plug-and-play framework for adaptive horizon execution that keeps the base VLA frozen. Given a fixed-length action chunk, its continuation head decomposes execution-horizon selection into a sequence of continue-or-replan decisions, which imposes an ordinal, prefix-sharing inductive bias over candidate horizons rather than treating them as independent classes. Since the optimal horizon for each chunk is not observable, we train this head with reinforcement learning from trajectory-level outcomes and introduce a Replanning-Efficiency Reward that jointly rewards task success and efficient VLA usage, discouraging the policy from collapsing to unnecessarily short horizons. On RoboTwin 2.0 with LingBot-VLA as the base policy, BCP improves the average success rate by +11.08% on 13 low-success tasks and from 89.88% to 93.94% (+4.06%) across all 50 tasks. Although trained only under the Clean setting, BCP generalizes to the Randomized setting, raising the average success rate by +4.06%. It also transfers to a different base policy $π_{0.5}$, achieving a better result on LIBERO (+1.7%) and, notably, on the harder LIBERO-PRO (+6.8%). On a real robot, BCP lifts success from 74% to 92% and from 44% to 84% on two manipulation tasks. Meanwhile, its negligible overhead, combined with higher success, makes BCP's overall runtime even lower than the fixed-horizon baselines.

The performance of Support Vector Machines (SVMs) critically depends on the kernel function choice, which enables implicit mapping of data into high-dimensional feature spaces. While classical kernels like Radial Basis Function (RBF) remain popular, orthogonal polynomial kernels offer mathematically interpretable alternatives that can incorporate structured prior knowledge. This work extends the orthogonal polynomial kernel paradigm by introducing a novel family based on discrete $q$-Hermite I polynomials, a class of $q$-orthogonal polynomials that generalize classical Hermite polynomials through a deformation parameter $q$. We formally define the q-Hermite kernel and establish its validity under Mercer's theorem. The kernel's inherent boundedness properties naturally prevent annihilation and explosion effects without requiring explicit scaling mechanisms. Extensive experiments across 20 benchmark datasets demonstrate that the proposed kernel achieves competitive performance compared to both classical kernels and other orthogonal polynomial kernels, while offering advantages in numerical stability and computational simplicity. Our results confirm that $q$-orthogonal polynomials constitute a promising direction for kernel design, bridging mathematical elegance with practical machine learning applications, that provides conceptual and algorithmic resources that may be further extended to emerging quantum computing paradigms. To facilitate full reproducibility, we provide the complete implementation and experimental pipeline in an open-access GitHub repository at https://github.com/Kokechacho/SVMs-QSVMs.

The adoption of large pre-trained multilingual models for neural machine translation (MNMT) faces a major challenge: excessive memory and computational consumption due to overly large vocabularies and embedding layers. Although existing compression methods like pruning, quantization and knowledge distillation reduce parameter redundancy, they mainly preserve the structure of the original vocabulary, thereby leaving a major source of inefficiency unresolved. We propose in this paper a general optimization framework that combines a vocabulary pruning method with a targeted fine-tuning protocol for MNMT models. We evaluate the proposed framework using three models (M2M100, NLLB-200, mBART-50) on the English-Arabic language pair. Our approach reduces the vocabulary size from over 128,000 to approximately 10,000 tokens, enabling a 60% memory saving without any loss in performance. Results show that optimized multilingual models can match or exceed the performance of dedicated bilingual baselines. In particular, the pruned and fine-tuned M2M100 model achieves a competitive BLEU score of 42.04 (against 44.59 for the OPUS-MTen- ar bilingual model) while it significantly outperforms it on the COMET metric (0.8730 vs 0.7911) revealing superior semantic adequacy and fluency.

Multimodal intent recognition combines linguistic, acoustic, and visual evidence, but individual modalities may be noisy, missing, semantically conflicting, or disproportionately dominant. Existing methods typically infer modality importance implicitly and either reweight or suppress unreliable inputs, without determining whether a degraded modality can be repaired and subsequently trusted. We propose PRIME (Precision-weighted Reliability Inference and Modality rEstoration), a closed-loop reliability guided framework that jointly diagnoses, restores, and reassesses modality quality at the sample level. PRIME represents the weakness of each modality through a contextual log-variance estimated from complementary diagnostic evidence, including predictive confidence, epistemic disagreement, cross-modal consensus, and feature degeneracy. Because modality-reliability annotations are unavailable, the estimator is explicitly trained using controlled modality corruption with known degradation severity, together with a heteroscedastic uncertainty objective. Rather than directly discarding an unreliable modality, PRIME uses its estimated weakness to control a prototype-conditioned variational restoration module that reconstructs the degraded representation from complementary modalities. Crucially, reliability is re-estimated after restoration, allowing the model to determine whether the repaired representation has become sufficiently trustworthy to contribute to prediction. The resulting post-restoration precisions are used for inverse-variance multimodal fusion. Experiments on multimodal intent-recognition benchmarks show that PRIME maintains competitive clean-data performance while improving robustness under missing, noisy, conflicting, and modality-imbalanced conditions.

Multimodal large language models (MLLMs) have made significant progress in chart understanding, generation, and editing, but their ability to annotate existing charts remains underexplored. Annotating charts is a common yet challenging communicative task, requiring models to infer intended messages, interpret chart semantics, and place appropriate textual or graphical elements. To address this gap, we introduce ChartAnno, a benchmark for evaluating MLLMs on chart annotation generation. It contains 1,200 real-world charts with paired code and annotation instructions across three levels of instruction specificity. We evaluate 10 representative MLLMs under two primary input settings: (1) chart code alone and (2) both chart code and chart image, and further include a chart image-only ablation study. Results show that proprietary models remain stronger overall, although large-scale open-source models narrow the gap. More specific instructions improve annotation quality, while inferring abstract intent remains most difficult for current MLLMs. Providing chart images brings limited overall gains, with improvements mainly appearing in design-related metrics. These findings highlight chart annotation generation as a challenging task requiring semantic grounding and effective annotation design. Code and data will be released in a future version.

While human evaluation is the gold standard in many NLP tasks, it suffers from prohibitive costs and poor scalability. When identifying top-performing models, typical evaluation protocols waste effort by exhaustively evaluating all models on the entire benchmark, a safe but inefficient approach. In this work, we formalize multi-model human evaluation as a best-arm identification problem in a multi-armed bandit setup with correlated arms, where pulling an arm corresponds to human-evaluating a model. By sampling adaptively based on the intermediate model rankings obtained on the samples so far, we can focus the annotation budget on the most competitive models. We prove the optimality of the proposed algorithms and show that it improves discrimination between top-performing models. This makes evaluations faster, cheaper and more aligned with large-scale competition evaluation goals.

Refurbishment-based noisy-label learning mixes an observed label with a model-derived pseudo target, typically using one sample-wise cleanliness score to control both branches. This creates a hidden coupling: reducing trust in the observed label automatically increases trust in the pseudo target. We show that this complementarity can replace one unreliable signal with another because a pseudo target learned from corrupted supervision may reproduce the noise it is meant to correct. Our representation diagnostics provide a consistent account of this mismatch: noisy supervision redirects deeper layers more strongly, whereas shallower relations remain comparatively stable and provide information beyond the loss posterior. We therefore propose TRACE, a Two-Source Reliability Assessment framework for Label Correction and Sample Reweighting. TRACE assesses the observed label using loss fit, shallow relation stability, and prediction agreement, while separately assessing the pseudo target using model confidence. Its source-specific scores control target correction and supervision strength without assuming complementary reliability. Across synthetic and real-world noisy benchmarks, TRACE improves representative refurbishment baselines and yields more reliable pseudo supervision.

We introduce the Infinite SLAM Transformer (SLAMFormer-$\infty$), the first geometric transformer capable of supporting both long-range frontend and backend processing without an explicit distance bound. Instead of relying on a first-frame-anchored formulation, SLAMFormer-$\infty$ employs memory conditions to define flexible coordinate systems and scales for input frames, enabling more expressive structural conditioning. Built upon this formulation, the frontend preserves efficient local computation, while the backend jointly optimizes long-range trajectories and scene geometry in a globally consistent manner. Experimental results demonstrate that SLAMFormer-$\infty$ achieves superior or highly competitive performance in both trajectory estimation and scene reconstruction across large-scale datasets. Notably, SLAMFormer-$\infty$ generalizes to extremely long trajectories, successfully operating on sequences exceeding $17\mathrm{km}$.

Large language models (LLMs) are now regularly asked to forecast real-world events, but comparisons are often difficult because models receive different information, use different tools, and are evaluated under different rules. This paper reports the completed \emph{AI World Cup} benchmark, in which ten LLM-based assistants made a single pre-tournament forecast of the entire 2026 FIFA World Cup. Every submission used the same tournament snapshot, prompt, JSON schema, and scoring procedure. The forecasts covered group-stage scores, group rankings, the knockout bracket, final placings, confidence values, and short explanations. After all 104 matches had been played, GPT-5.5 Thinking finished first with 744 points, followed by GPT-5.5 with 717, Gemini with 699, and Qwen 3.7 with 687. GPT-5.5 Thinking was also the only model to select Spain, which defeated Argentina 1--0 in the final, as champion. The final ranking was driven mainly by knockout performance: total score was strongly correlated with knockout points ($r=0.986$), but showed little relationship with group-stage match points ($r=0.055$), group-standing points ($r=-0.103$), or their combined pre-knockout score ($r=-0.054$). Match-level accuracy produced a different ordering. Claude Sonnet 4.6 correctly predicted the largest number of group-stage outcomes (63.89\%) but placed sixth overall. Average self-reported confidence was also unrelated to either outcome accuracy ($r=-0.060$) or total score ($r=-0.067$). The results suggest that forecasting a complete tournament tests something different from predicting matches one at a time, while also showing how strongly a bracket-based leaderboard can depend on scoring design. The benchmark materials, raw responses, and scoring code are released to support replication and future extensions.

Accurate Uncertainty Quantification (UQ) is critical for reliable deployment of Large Language Models (LLMs), yet traditional probability-based metrics often fail to capture the model's true epistemic state. While recent mechanistic approaches leverage hidden state dynamics, they typically aggregate residual stream updates, conflating the distinct roles of parametric memory (Feed-Forward Networks) and contextual processing (Attention). We argue that this aggregation obscures fine-grained mechanistic conflicts, such as memory-context misalignment, that are fundamental indicators of uncertainty. To address this, we introduce \textbf{D}ecoupled \textbf{U}pdate \textbf{D}ynamics \textbf{(DUD)}, a framework that explicitly decouples FFN and Attention contributions via noise-induced causal interventions. By quantifying the independent restoration capabilities of each module, we construct a dual-stream dynamic profile that captures the model's internal fragility. Extensive experiments demonstrate that DUD significantly outperforms state-of-the-art baselines in both uncertainty estimation and calibration, while exhibiting superior cross-dataset generalization, validating decoupled dynamics as a robust proxy for model faithfulness.

Road network segmentation from satellite imagery remains challenging due to large geographic variation in road appearance, occlusions, and domain shifts introduced by differing resolutions and sensors. Existing models, typically trained under narrow resolution--region combinations, generalise poorly to unseen environments such as rural settings, regions with distinct road materials, or imagery from new satellite platforms, often producing broken or disconnected predictions. Adapting these models to new domains usually requires retraining or fine-tuning, which is costly and risks catastrophic forgetting. In this work, we reframe global road extraction as a continual adaptation problem rather than an architectural one. Our framework combines cross-resolution knowledge distillation across a resolution-decreasing curriculum, multi-sensor training, and topology-aware supervision, yielding a single model that generalises across $0.3-1.0$ m imagery from multiple satellite platforms across continents. On publicly available benchmarks, including City-Scale and Global-Scale, our model outperforms state-of-the-art results by up to $22$ F1 points and $15$ APLS points, while remaining the most efficient, with $3\times$ faster inference. Our results suggest that improved robustness across diverse sub-meter satellite imagery can be achieved through targeted training strategies, such as data curricula, distillation, and topology-aware losses, rather than increasingly complex architectures.

Precise anomaly localization over long-context time series is a crucial task in monitoring applications across clinical care, industrial operations, financial services, and logistics, where brief evidence may hide inside long spans of high-frequency data. Time-Series Language Models (TSLMs) are able to ingest time series data and verbalize findings on anomalies in natural language; however, recent benchmarks report a decrease in retrieval performance at long contexts, mirroring failure modes in text, vision, and audio. In the text domain, Recursive Language Models (RLMs) can recover much of this lost performance by keeping context external to the large language model (LLM), allowing the model to query it through code. We present TimeRLM, an RLM formulation for time-series that sequentially manipulates the signal using code and vision capabilities. We further introduce AnomalyXL, a synthetic long-context anomaly localization benchmark with programmatically injected anomalies that require precise retrieval. We implement five different task categories and two variants: AnomalyXL-MCQ and AnomalyXL-Localize. TimeRLM outperforms every evaluated TSLM and single-pass baseline on four of the five AnomalyXL-Localize tasks, reaching 0.682 IoU on localization and 0.745 on classify-with-evidence, versus at most 0.329 and 0.072 across all baselines. We post-train TimeRLM using reinforcement learning. The resulting model further improves performance and requires approximately one-third as many agent interaction turns as its untrained base model to produce a final answer. On unseen real-world ECG, sleep and software observability recordings, the post-trained TimeRLM retains or improves performance, surpassing TSLMs despite being trained exclusively on synthetic data. Our findings suggest recursive interaction with time-series is an effective approach for long-horizon retrieval.

Humanoid robots have the potential to perform dexterous manipulation in human environments, yet acquiring diverse and generalizable skills remains costly due to expensive hardware data collection and labor-intensive annotation. Recent advances in video generative models provide a promising opportunity to synthesize rich manipulation experiences from visual observations, but transferring such imagined behaviors into executable whole-body humanoid skills remains largely unexplored. In this work, we present RoboReact, a framework that automatically synthesizes whole-body humanoid manipulation skills from a single egocentric RGB-D observation. RoboReact generates human manipulation videos, extracts geometry-preserving interaction keyframes through depth-aware 3D reconstruction, and retargets them to high-DoF humanoid platforms while preserving hand-object interaction geometry. To bridge the gap between imagined plans and physical execution, RoboReact performs online object-centric re-grounding and leverages a vision-language model-guided refinement loop to adapt skills under geometric mismatch and execution deviations. The refined skills are executed through a whole-body controller, enabling coordinated whole-body manipulation and dexterous interaction. Experiments on real humanoid robots demonstrate that RoboReact generalizes across diverse object configurations and robustly recovers from execution disturbances without requiring teleoperation or human demonstrations. These results highlight the potential of combining generative models, vision-language reasoning, and closed-loop control for scalable humanoid skill acquisition.

The development and testing of advanced aerial robots require experiments in controlled environments with tailored airflow profiles. This paper presents an online learning algorithm for controlling the complex airflow field in a multi-fan vertical wind tunnel. Our method combines a simplified physical model with iterative, measurement-based learning, enabling sample-efficient convergence to desired airflow distributions. We demonstrate the method's versatility by generating complex airflow, such as uniform, Gaussian, and parabolic profiles. Crucially, we show that our algorithm can produce an airflow profile specifically designed for passive soaring, greatly enhancing flight performance of a soaring robot. Variability, practical utility, and robustness of our approach are further highlighted by successful operation with a varying number of fans.

AI systems rewrite information constantly: conversations become stored memories, documents become answers. The rewrite can keep a claim while washing away what made it checkable, who said it, how sure they were, when it held. We call that failure factwashing, and release factwash, an open-source write-time gate that catches it deterministically, with named flags and evidence rather than an LLM judge. Building it answers a practical question: when does a cheap check suffice, and when do you need a model? What decides is whether the property has a bounded surface-cue inventory. Explicit negation cues are close to enumerable, so a word list finishes and transfers, reaching 0.91 F1 on untuned text. Hedging and attribution have open-ended realizations, so vocabulary plateaus near half recall, and a one-question LLM witness recovers +17 and +15 points of cue-detection recall at equal precision. Deployed, that witness may only lower a verdict, so it buys precision rather than coverage. We measure cue detection on 105,596 independently annotated sentences. A blind-labelled corpus of memory writes then locates the failure: 55% of bad writes in conversational hearsay, 7% in business email (p < 0.001), so the first deployment question is not which detector to use but whether the failure occurs at all. On unmodified mem0 2.0.7, the gate flags 5 of 8 hedged-hearsay writes.

Non-intrusive reduced-order models (NIROMs) have become a standard tool for approximating parametric partial differential equations from computer design of experiments while significantly reducing computational costs. However, assessing the reliability of their predictions remains a major challenge, particularly in extrapolation regimes or under limited training data. In this work, we introduce a framework for quantifying model-form uncertainty in NIROMs by combining a perturbative stochastic representation of reduced bases with distribution-free conformal-type methods. Starting from a deterministic reduced basis constructed from snapshot matrices, we model uncertainty through random perturbations defined on the Stiefel manifold, directed along the discarded modes, yielding stochastic reduced-order approximations whose induced variance reflects the basis-truncation error. A transport approximation gives a closed-form posterior variance that sepa- rates basis-induced from regression-induced uncertainty, without re-training the underlying Gaussian processes. We include this posterior variance within a conformal risk control calibration framework, that provides prediction sets with coordinate miscoverage guarantees. The calibration factor produced by this framework is itself an interpretable, scalar diagnostic of the quality of the uncertainty estimate. The methodology is evaluated on parametric PDE benchmarks and an industrial tire-manufacturing calendering process. Numerical experiments demonstrate reliable, locally informative uncertainty quantification that goes beyond the Gaussian predictive variance.

Existing visual emotion understanding methods typically ignore cultural variations in emotional perception. We introduce culture-conditioned visual emotion understanding, a task that predicts the culture-specific emotional perception of a given image and explains the underlying rationale. Although related benchmarks exist, they are limited by inconsistent individual annotations, which hinder the derivation of majority-supported culture-level emotion labels, and imbalanced cultural coverage. Thus, we present ArtECulture, a benchmark containing 6,792 artworks with culture-specific emotion labels and explanations across English, Chinese, and Arabic cultures, with balanced Western and non-Western content. Evaluations of 16 open- and closed-source Multimodal Large Language Models (MLLMs) under a zero-shot setting reveal that the task remains challenging, with the best model achieving below 50\% accuracy. To address this limitation, we introduce a retrieval-augmented culture-conditioned emotion understanding framework, which leverages a concept-based cultural emotion knowledge base to inject explicit cultural knowledge into MLLMs without additional training. The framework improves both culturally aligned emotion prediction and grounded explanation generation. Our benchmark and code will be publicly released.

Predicting LLM's capabilities on real-world tasks is essential, yet the extent to which performance on commonsense benchmarks predicts downstream performance remains underspecified. To establish the practical usability of widely adopted commonsense benchmarks, we evaluate 23 models from six families on four established commonsense benchmarks, four reworked variants, three non-commonsense controls, and eight downstream tasks requiring implicit social, pragmatic, temporal, or physical reasoning. We compare model rankings, compute controlled correlations, and use leave-one-family-out cross-validation to assess the criterion validity of commonsense benchmarks. Our results show that revised benchmarks largely preserve original model rankings and do not improve downstream predictive power. Commonsense benchmarks show consistent cross-family predictive validity for only a narrow subset of downstream tasks, with smaller or metric-specific gains elsewhere. Overall, standardized commonsense benchmarks provide task-dependent rather than broad evidence of downstream commonsense competence.

Federated learning enables collaborative model training across distributed edge devices while strictly preserving data privacy. To facilitate practical deployment on resource-constrained edge devices, Spiking Neural Networks (SNNs) have emerged as a promising alternative to traditional Artificial Neural Networks (ANNs) due to their sparse computing mechanisms and high energy efficiency. However, jointly training ANNs and SNNs exposes a challenge of representational misalignment, which is intrinsically caused by differences in information representation, specifically the semantic gap between continuous real-valued activations in ANNs and discrete spatio-temporal spikes in SNNs. To overcome this barrier, we propose AS-FedBridge, a novel federated learning framework tailored for mixed ANN-SNN clients. AS-FedBridge features a lightweight Bridge equipped with a Pseudo-Spike Interface, which effectively projects continuous signals into a spike-compatible space to facilitate ANN-SNN alignment. Given the absence of existing mixed ANN-SNN federated frameworks, we establish a comprehensive benchmark to evaluate against multiple advanced heterogeneous FL methods. Our empirical analysis demonstrates a positive correlation between the degree of ANN-SNN alignment and the collaborative FL performance. Across four datasets, AS-FedBridge consistently demonstrates advanced accuracy while mitigating extreme scale, architecture, and client heterogeneity challenge. Furthermore, our framework enables a highly controllable trade-off between model performance and resource efficiency. AS-FedBridge accomplishes these robust performance gains while introducing only marginal computational overhead, establishing a robust and practical foundation for mixed ANN-SNN federated learning systems.

Future integrated sensing and communication (ISAC) architectures separate the sensing entity (SE) that acquires measurements from the sensing function (SF) that performs inference, creating a need for compact, task-oriented feedback on the SE-SF interface. Forwarding the raw channel frequency response or full per-link delay-Doppler-azimuth-elevation (DDAE) tensor is prohibitively expensive, while peak-only reporting discards target-discriminative structure under clutter. We propose a learning-based coarse-to-fine sensing pipeline with candidate-latent feedback for single-target estimation. At the SE, a lightweight convolutional scorer produces a dense delay-Doppler proposal map from pilot-based OFDM channel estimates, and a learned encoder constructs K compact C-dimensional candidate tokens by fusing per-candidate azimuth-elevation patches, normalized position, and confidence cues. The latents are uniformly quantized post-training to b bits and transmitted under a finite budget B_fb = bKC + 18K + 16 bits to the SF, which performs cross-candidate refinement, reranking, and joint four-parameter estimation. On a ray-traced urban scene with static and dynamic clutter, three operating points in the (K, C, b) design space achieve 96.33-98.88% detection at 107-806 bytes per coherent processing interval, compression ratios of 1.2-9.2 x 10^4 over the 8-bit DDAE magnitude tensor, reducing the SE-SF interface from multi-Gbit/s to sub-Mbit/s rates. Cross-scene evaluation on an independent campus-scale environment achieves 98.79-99.50% detection and at-or-better angular accuracy without retraining, indicating that the learned representation captures target-relevant structure that transports across scenes of comparable or lower clutter density.

A standard claim in the literature on retrieval-augmented and memory-augmented language models is that shorter context is better when the relevant information is preserved. We test this claim by running every sample of two long-context benchmarks -- BABILong and GraphWalks (BFS) -- at four context-retention fractions (100%, 75%, 50%, 25%) under two truncation protocols. The first is the naive protocol implicitly used in much prior work: drop content from the middle of the prompt. The second is distractor-aware: identify the task-relevant content for each sample and drop only the rest. We evaluate three sizes of the Claude family (Haiku 4.5, Sonnet 4.6, Opus 4.7) and, to test cross-provider generality, GPT-5.5 from a different provider; we apply the same protocol to two further benchmarks (MRCR v2, Oolong). Under naive truncation, score collapses monotonically (paired Wilcoxon, Holm-corrected p_adj < 0.05 in all eight BABILong and GraphWalks cells). Under the distractor-aware protocol -- which preserves the signal by construction -- performance is preserved or improves: the two smaller Claude models show statistically significant gains on BABILong, while the larger models (Opus 4.7 and GPT-5.5) sit at their full-context ceiling. The naive collapse and its distractor-aware recovery replicate on GPT-5.5, ruling out a single-provider artifact. The mechanism is direct: under the naive protocol the answer-bearing content survives in fewer than 1% of samples at 25% retention; under the distractor-aware protocol it is preserved by construction. The naive protocol is therefore not a measurement of context-window effects; it is a measurement of how often middle-removal happens to spare the answer. We conclude that future studies of context-length effects must specify how they distinguish signal from distractor, or they are at best ambiguous between two opposite hypotheses.

Accurate extrinsic rotation calibration between sensors is fundamental to the performance of robotic perception systems. However, most existing calibration techniques rely on full 6-DoF motion to excite all degrees of freedom, which is often infeasible for ground-constrained robots with limited motion capabilities. Recent approaches designed for such restricted settings, such as Canonical Correlation Analysis (CCA)-based methods, suffer from ill-conditioned covariance matrices that lead to numerical instability and suboptimal calibration accuracy. To overcome these limitations, we present a novel rotation calibration framework named PLS-Calib that, for the first time, leverages Partial Least Squares (PLS) regression to model the latent kinematic correlations between asynchronous, heterogeneous sensor streams. Specifically, we apply our method to the calibration of an event camera and an odometry onboard a ground robot. To improve event-based pattern detection, we introduce a polarity-aware event representation, which enhances spatiotemporal contrast in circular calibration targets. Our PLS-based formulation yields a closed-form, stable solution that avoids matrix singularities inherent in CCA-based approaches. Extensive experiments on both synthetic and real-world datasets validate the effectiveness of our approach, demonstrating significant improvements in calibration robustness and accuracy over state-of-the-art methods. This work offers a practical and theoretically grounded solution for rotation calibration in constrained robotic systems and opens up new directions for applying statistical learning techniques in neuromorphic vision.

Recent data-driven methods for synthesizing 6-DoF grasp poses use generative models to learn complex grasp pose distributions and generate diverse candidate poses. In particular, SE(3)-equivariant flow-based models generate grasp poses that transform consistently with object rotations and translations. However, these methods sample by iterative numerical integration, requiring tens of function evaluations per grasp and limiting their use in real-time manipulation. We propose GraspMeanFlow, an SE(3)-equivariant MeanFlow framework for few-step 6-DoF grasp generation. Our method learns the average velocity over a finite time interval, defined through the time-ordered exponential so that it reproduces exactly the rigid-body displacement accumulated over that interval. We prove that a point-cloud-conditioned distribution transported by an equivariant average-velocity flow map remains invariant, so equivariance is retained under few-step sampling, and we condition the field on a pair of times by lifting both to equivariant vectors, leaving the backbone otherwise unchanged. For stable training, we pair a flow-matching boundary term with either of two consistency terms: the differential MeanFlow identity, whose target requires a Jacobian-vector product, or an equivalent semigroup loss that avoids it. Experiments on ACRONYM show that a single function evaluation of GraspMeanFlow reaches the EMD that an iterative SE(3) flow model needs five steps to approach, that a second instantiation of the same framework improves grasp success by up to 24.3 points in the few-step regime, and that both generate grasp distributions transforming exactly with the object.

Differentially private zeroth-order optimization (DP-ZO) enables memory-efficient private fine-tuning of large language models using only forward evaluations. Existing aggregation-based DP-ZO methods reconstruct model updates at a fixed scale, ignoring that the strength of useful signals varies throughout training. Consequently, noise-dominated updates may receive excessive weight and degrade model utility. To address this issue, we propose SAGE, a noise-aware shrinkage method that adaptively attenuates privatized estimates according to their estimated signal quality. SAGE subtracts the known Gaussian noise variance from the observed second moment to estimate the underlying signal energy, stabilizes this estimate through temporal tracking, and compares its current signal-to-noise level with a warm-up reference to derive a bounded shrinkage factor. As pure post-processing, SAGE requires neither additional privacy budget nor model queries and introduces only constant additional state. Our theoretical analysis shows that shrinkage reduces the quadratic update-risk term faster than the linear descent term, preserving useful descent while limiting the influence of noise-dominated updates. Experiments on RoBERTa-large, OPT-1.3B, and OPT-6.7B demonstrate that SAGE outperforms existing baselines in most settings under the same privacy budgets while preserving the forward-only memory efficiency of DP-ZO.

Parameter-efficient fine-tuning (PEFT) enables efficient adaptation of large language models, but existing MoE-based PEFT methods typically improve capacity by storing multiple full LoRA experts, causing adapter storage to grow linearly with the number of experts and restricting adaptation to a fixed expert pool. We ask whether MoE-based PEFT can produce instance-specific adaptations without explicitly storing a separate LoRA module for each expert. To address this gap, we propose MoEGen, an adaptation framework that shifts MoE-based PEFT from expert selection to expert-conditioned parameter generation. Instead of storing each expert as a full LoRA adapter, MoEGen represents each expert as a small learnable vector, termed an expert code. It routes each input over these vectors and uses their weighted combination to condition a lightweight hypernetwork that generates input-specific low-rank updates. This design decouples expert capacity from adapter storage while enabling instance-conditioned adaptation. Experiments on eight commonsense reasoning benchmarks show consistent improvements over strong static and MoE-based PEFT baselines across three backbones. MoEGen also performs strongly in joint medical and legal-domain adaptation.

As time-series foundation models have emerged, the need for benchmarks that can evaluate their forecasting ability in meaningful ways has become increasingly important. Existing time-series forecasting benchmarks provide useful standardized comparisons, but they often evaluate heterogeneous series with uniform error-based metrics. Strong performance under such metrics does not necessarily imply that a model's forecasts will support the best real-world decisions across domains. For example, in stock forecasting, correctly predicting whether a price will rise or fall can be more directly relevant to realized returns than minimizing point-wise forecast error alone. To this end, we introduce FinVerse, a finance-domain time-series forecasting benchmark that takes a first step toward more realistic evaluation. The released FinVerse data artifact contains 116,897 financial time series with 171.1M observations, of which 60,232 series with 17.4M observations are selected as evaluated targets based on their economic relevance to financial decisions. Unlike generic forecasting benchmarks that primarily emphasize uniform point-forecast or probabilistic accuracy, FinVerse defines 11 metric families comprising 78 evaluation metrics and assigns the most appropriate evaluation metrics to each individual time series based on its underlying economic meaning. Our analysis of 43 public time-series forecasting foundation models shows that strong performance under generic forecasting criteria does not necessarily translate into useful financial forecasts. This finding highlights the need for domain-aware benchmarks that evaluate models under objectives closer to real-world decision making.

Large language model-based multi-agent systems (LLM-MAS) are designed through roles, debate protocols, and aggregation rules. These choices create implicit social expectations: agents may be expected to trust, challenge, defer to, or collaborate with peers. We study the effects of making inter-agent relation semantics explicit. We use a minimal signed-network formulation of relational priors and inject natural-language renderings into agent system prompts while holding the task protocol fixed. Across a commons-governance simulation and multi-agent debate, relational priors primarily act as convergence pressure: increasing relational positivity tends to make agents coordinate or agree more readily. This pressure can help when utility rewards behavioral alignment, as in sustainable resource governance and subjective consensus. It does not, however, reliably improve accuracy. In objective QA debates, higher positivity can increase agreement even when correctness-conditioned agreement does not improve and may decline in some settings. Effects vary by model backbone, relation type, and topology; explicit neutrality is not equivalent to omitting relational framing. We argue that relational priors should not be a default add-on for LLM-MAS. Their safer use is diagnostic and task-specific: compare against a no-prior baseline, monitor correctness-conditioned metrics when truth matters, and omit the relational layer when validation does not justify it.

Motion priors provide powerful guidance for learning naturalistic humanoid behaviors. However, existing methods typically learn a general, task-agnostic prior from the entire reference dataset and apply it uniformly throughout policy training. As a result, the prior cannot distinguish which reference motions are relevant to the current task context, potentially providing irrelevant or conflicting guidance. We present Context-Aware Motion Priors (CMP), a framework that adapts a general motion prior to the current task context without manual skill labels, dataset partitioning, or a separate skill discovery stage. Specifically, CMP learns context-motion compatibility using high-advantage policy rollouts, while a demonstration-based objective keeps the learned relevance grounded in the reference distribution. The resulting relevance scores reweight reference supervision for training a lightweight context-conditioned adapter. To evaluate the effectiveness and generality of CMP, we instantiate it with both Adversarial Motion Priors and Score-Matching Motion Priors. Across five humanoid control tasks, CMP consistently improves task performance and sample efficiency, learns meaningful context-motion alignment, and remains robust to imbalanced reference distributions. These results show that adapting motion priors to task contexts provides more relevant guidance for humanoid policy learning.

Vision-Language-Action (VLA) policies promise general robotic manipulation, but their robustness against physical-world attacks remains fragile. In particular, we show that physically realizable adversarial patches can reliably induce failures by triggering a mechanism we call policy-critical action-to-vision attention hijacking, where action-conditioned attention is diverted from task-relevant regions to a localized patch. To demonstrate the threat, we propose Attention-Guided Semantic Disruption (AGSD), an Expectation-over-Transformation (EOT) optimized printable patch that jointly (i) concentrates action-to-vision attention on the patch and (ii) disrupts vision-language semantic alignment, yielding strong cross-task and cross-architecture transfer. To mitigate such attacks, we introduce Structure-Aware Robust Fine-Tuning (SARF), a zero-inference-overhead defense that fine-tunes only the visual encoder using feature anchoring, policy-critical attention correction, and language-guided geometric consistency restricted to semantically relevant regions. On LIBERO, SARF reduces OpenVLA's failure rate under AGSD from 100% to 14.2%-56.8% (28.6% average) across suites while preserving clean performance, and on a real PiPER manipulator it improves average success under AGSD from 23.0% to 65.0%. These results highlight mechanism-level robustness as a practical path to securing VLA robots against physical attention hijacking.

Existing low rank KV cache methods preserve either model weights or key variance, neither of which directly reflects the attention scores used during inference. We derive the expected attention score distortion caused by rank r key compression and show that it yields a covariance weighted low rank objective. Under a margin condition, controlling this distortion also improves top k recall. The optimal rank r solution has a closed form asymmetric factorization obtained from the SVD of the covariance weighted query key operator. This motivates SAKI, a training free KV cache index that directly preserves attention scores rather than key reconstruction quality. Across LLaMA 3.1 8B, Qwen 2.5 7B, Mistral 7B v0.1, and Llama 3.2 3B, SAKI outperforms key PCA at every tested rank. At rank 32, it removes 13 to 30 percent of PCA's remaining top 64 recall error, including improvements from 0.748 to 0.799 on LLaMA 3.1 8B and from 0.786 to 0.850 on Qwen 2.5 7B. It improves 68 to 89 percent of attention heads per model, with the largest gains in deeper layers. Predicted score MSE reductions closely match empirical measurements, with a Pearson correlation of 0.997, while ablation studies confirm that the gains arise from optimizing the attention score objective rather than covariance weighting alone. Analysis of the scoring operator further explains why weight only, invariant subspace, and key reconstruction methods can be suboptimal.

Motion priors improve reinforcement learning for physics-based humanoid tracking, but temporal priors require ordered motion clips, while pose priors provide limited guidance for policy-induced pose transitions. We present Pose Flow Matching for Humanoid Robots (PFM-HR), a reusable flow matching prior trained directly on large scale unordered pose data. PFM-HR introduces the Pose Geometry Score (PGS), which quantifies how joint coordinate changes during rollouts align with the local geometry of pose variation captured by the prior. Using PGS to modulate the tracking reward guides policy exploration toward structured pose changes while keeping the prior frozen across tracking tasks. Experiments demonstrate that PFM-HR improves both single motion and general motion tracking, especially for highly dynamic motions.

Local Flexibility Markets (LFMs) require Distribution System Operators (DSOs) to determine both the quantity of flexibility to procure and the corresponding willingness to pay during market clearing. Existing approaches typically rely on unrealistic centralized AC-OPF clearing algorithms or strictly localized, static flexibility requests driven primarily by congestion management, while the economic value of flexibility is largely neglected. This paper proposes a dynamic flexibility-request methodology in which the DSO's willingness to pay is embedded directly into the market-clearing objective by monetizing transformer and cable aging, network losses, and voltage congestion. To enable computationally efficient clearing, exact convex piecewise-linear epigraph reformulations of the IEEE C57.91 transformer aging model and an Arrhenius-based cable aging model are developed and formally proven to preserve exactness. The proposed framework is validated on a modified CIGRE MV benchmark and compared with a recent state-of-the-art flexibility-request methodology. The results demonstrate significantly higher market liquidity, more efficient flexibility procurement, and improved network operations while preserving the market's transparency and non-discrimination principles.

Agentic reinforcement learning enables LLM agents to learn through interaction, but sparse trajectory-level rewards reveal success without identifying which intermediate decisions deserve credit. Training-only privileged skills can provide denser supervision by allowing the same frozen policy snapshot to rescore fixed tokens from skill-free trajectories while conditioned on task-matched procedural skills. Existing methods, however, do not jointly calibrate teacher scores across interaction steps, relate teacher confidence to realized returns, and integrate the resulting signal into native reward-to-advantage construction. We introduce Agentic Reinforcement Learning with Self-Distilled Reward Shaping (ADRS), a framework for constructing return-associated token-level credit for multi-turn language agents. ADRS centers and normalizes privileged token scores within each step, modulates them with a return-associated Teacher Value Advantage (TVA) gate based on within-group confidence--return association, and incorporates the gated token signal into native RL credit construction. Together, these components determine what the teacher prefers, when that preference is return-relevant, and how it enters the native reinforcement-learning credit path, while keeping rollouts and inference skill-free. Finally, experiments across three interactive benchmarks show that ADRS consistently improves performance on long-horizon tasks, with gains persisting across RL backbones, reduced-data settings, unseen tasks, and extended training. For anonymous review, our code is available at the following the link: https://github.com/gitrxh/ADRS-arxiv

Benchmark gains are often treated as evidence of greater LLM capability. Yet the same gain can reflect different changes in model behavior. A model may reach new answers, or produce answers that were already within reach. Aggregate scores do not distinguish these changes question by question. We establish a question-level audit under fixed budgets, temperatures, and answer formats. A question is realized when the default deployment procedure produces the correct answer. A question is reachable when a specified probe finds that answer within a fixed budget. We first test whether inference-time layer routing can expand reachability. Under a matched budget, random routes match or exceed structured search in all 43 model and task settings. Answer-blind procedures retain almost none of this gain, which instead requires access to the correct answer. We then ask why reachable answers sometimes fail to appear. Across six cases spanning 0.5B to 31B, silencing one identified MLP block repairs 68 to 92 percent of a predefined failure set. We next test whether training closes the gap by expanding reachability. In five of six matched evaluations, deployed performance rises while the reachable ceiling remains flat or falls. For DAPO, the deployed score rises by 14.7 points while the reachable ceiling falls by 13.3 points. Across the settings we audit, realization and reachability therefore do not always change together. Claims of capability expansion should report both realized performance and reachability under matched evaluation conditions. Code is available at https://github.com/LiZaiyuan0619/reachability-not-realization

Visual world models typically learn future dynamics from a single observation stream, limiting their ability to model cooperative systems with multiple independently moving observers. We investigate this challenge in Mother--Child endoscopic retrograde cholangiopancreatography (ERCP), where two flexible scopes provide complementary yet role-dependent views without a calibrated stereo relationship. Unlike conventional multi-view fusion that assumes symmetric information exchange, we formulate \textbf{role-asymmetric dual-scope future prediction}, where cross-view evidence is selectively transferred according to the prediction target and its underlying spatial requirements. We propose \textbf{CrossScope}, a dual-stream surgical world model that preserves view-specific experts while enabling target-specific evidence routing through geometry-guided residual interactions. CrossScope learns two complementary communication directions: geometric motion cues from the Mother view guide Child-view future dynamics, while pose-aligned Child appearance supports Mother-view prediction only when valid spatial correspondence is established. This design allows each scope to contribute task-relevant evidence without compromising its view-specific representation. To evaluate this problem, we establish a paired dual-scope benchmark comprising synchronized phantom and real-world ERCP episodes, with evaluations assessing visual fidelity, structural preservation, target localization, and motion consistency. Experiments demonstrate that CrossScope consistently outperforms strong surgical video generation baselines, validating the importance of role-aware evidence routing for multi-observer visual world modeling.

Flow-matching vision-language-action (VLA) models such as pi0 generate robot actions by integrating a learned denoising velocity field, and have been reported to resist adversarial perturbations that readily fool autoregressive VLAs. We show that this robustness is largely illusory: it stems from prior attacks ignoring the multi-step denoising ODE. We introduce DRIFT (Denoising Redirection via Input perturbation of the Flow-matching Trajectory), a test-time universal adversarial patch placed on the robot's gripper that attacks the denoising velocity field of an off-the-shelf policy. Our central finding is counterintuitive: attacking only the first denoising step is both stronger and cheaper than attacking a wider window of steps, which we explain through a gradient conflict unique to input-space optimization and which is exactly opposite to the training-time backdoor regime. On pi0 and pi0.5 across four LIBERO suites, DRIFT breaks essentially all originally-solvable tasks with a small single patch, far exceeding action- and embedding-space attack baselines.

Large language models (LLMs) power educational applications from tutoring to essay scoring, but each is a point solution to a single task, and only recently have these point solutions been integrated into agents operating over a learning management system (LMS). Yet tutoring is long-horizon, since a learner improves over days and weeks rather than in a single turn, and no benchmark evaluates an agent tutor across a sustained relationship. We introduce EduClaw-Bench, a benchmark that places an agent tutor in a continuous 30-day relationship with a simulated learner grounded in knowledge tracing (KT), whose knowledge-concept mastery, from a KT model trained on real-student data, drives its answers and is probed for learning gain across 55 scenarios. Each agent is scored on three primary axes (learning gain, responsiveness, and helpfulness) and two curriculum-design axes (Gagné and Rosenshine), with helpfulness and the curriculum axes judged by a cross-family panel of three LLM judges. Evaluating 10 agent adapters over three base-model tiers yields two findings that single-tier, single-session evaluation cannot reach. First, tutoring quality belongs to the base model and the agent harness together rather than either alone. Second, almost no combination sustains good tutoring over the full horizon. A calibration check ($\text{ECE}=0.049$) and a live-classroom field study confirm that the simulated learner and its measurements track reality. Our work is a step toward trustworthy AI tutors for future education.

Post-training reinforcement learning (RL) algorithms are commonly used to align large vision-language models (LVLMs) with human intent and the requirements of visual reasoning tasks. However, existing RL-based alignment methods are often resource-intensive and encounter mismatches between training objectives and inference-time distributions. To bridge this gap, we propose a novel test-time alignment approach that leverages trajectory-guided structured sampling for dynamic inference-time refinement, achieving better alignment with visual grounding and ensuring logical consistency. Our approach begins with curating a reasoning memory bank via a trajectory learning algorithm, which decomposes complex question solving into ordered sequences of predefined reasoning patterns. It subsequently accomplishes inference-time alignment by first collecting trajectories from reasoning memory bank to establish a global structural reasoning prior, and then using an iterative Markov Chain Monte Carlo (MCMC) algorithm for localized multi-objective refinement of the reasoning trace. Experiments across multiple multimodal reasoning datasets demonstrate that our approach significantly improves accuracy without incurring prohibitive inference overhead. These results establish trajectory-guided test-time sampling as a scalable and effective alternative to traditional post-training alignment, particularly for complex visual reasoning tasks.

Structure-preserving de-identification replaces protected health information (PHI) with realistic same-type surrogates -- "Anna S." becomes "Maria S.", not [NAME] -- so that clinical text stays fluent and downstream tools keep working. But this only helps if the substitution does not itself corrupt the signal those tools rely on. We ask a narrow, testable question: on the spans a de-identifier actually masks, can downstream PHI detectors still find the surrogate? We introduce a paired, multi-detector evaluation protocol that (i) scores utility only on masked spans, decoupling coverage from utility; (ii) uses equivalence testing (TOST) rather than null-hypothesis significance testing, which is uninformative at our sample size (57k paired spans); and (iii) builds a surrogate-failure typology separating fixable generator defects from intrinsic detector limits. Across 11 detectors, 7 benchmarks, and 7 languages (1,750 documents), recall on masked spans moves from 76.1% to 74.9% -- a change our equivalence test shows is statistically equivalent to zero within a +/-2-point margin (p ~ 3e-9), with detector ranking preserved. The residual loss does not reflect detectors getting worse at PHI: it concentrates in malformed and out-of-distribution surrogates (truncation Chicago -> Illino, salience loss Cedars-Sinai -> Vidant). A redaction floor and an open-source surrogate baseline indicate the effect is a property of well-formed substitution, not of one tool. We release the evaluation subsets, scoring code, and an interactive dashboard at https://custodianai.pages.dev so the protocol can audit any structure-preserving transform.

Lecture videos distribute knowledge across speech, slide text, diagrams, equations, and presentation order, which transcript-only retrieval does not fully preserve. This paper presents an evidence-grounded multimodal pipeline that transcribes lectures, selects semantic anchors, applies optical character recognition (OCR), and uses a vision-language model to extract only concepts and typed relationships supported by transcript, OCR, or visual evidence. Mentions are validated and canonicalized into a provenance-rich knowledge graph. On three neural-network lectures, the pipeline processed 3,118 frames, 756 transcript segments, and 559 anchors. It retained 1,022 concept and 312 relationship mentions, yielding 172 canonical concepts and 282 relationships with 90.38% endpoint coverage. A preliminary three question retrieval test achieved 100% top-1 and top-3 accuracy and 100% mean top-5 recall. The contribution is an auditable construction method rather than a state-of-the-art performance claim.

This research proposes a constrained motion planning framework for robot manipulators in human-robot interaction (HRI). For a non-redundant manipulator with a fully specified end-effector pose, additional requirements such as collision avoidance and self-collision avoidance are difficult to handle as simple null-space secondary tasks. This limitation makes it challenging to generate feasible joint-space trajectories in HRI environments where safety and kinematic constraints must be considered simultaneously. To address this limitation, collision- and self-collision-aware trajectories are generated using Rapidly-exploring Random Tree (RRT) and RRT* algorithms, and the resulting dataset is used to train a diffusion model that generates constraint-satisfying trajectories through guided sampling. To reduce the inference time required for iterative diffusion sampling, consistency distillation is applied, and a joint-weighted jerk regularization term is incorporated into the loss function to promote smoother trajectories by penalizing abrupt changes in joint acceleration. Simulation results show that the consistency model generates 150 trajectory candidates in less than 100 ms, maintains a high episode success rate, and substantially reduces joint and end-effector jerk when jerk regularization is applied.

Biomedical relation extraction (BioRE) extracts structured knowledge from biomedical literature for applications such as knowledge base construction and hypothesis generation. Traditional symbolic systems such as SemRep provide high precision but limited recall, while large language models (LLMs) offer stronger contextual reasoning but remain prone to false-positive predictions. We developed ANCHOR-RE, a framework that integrates ontology-guided reasoning, external knowledge grounding, and data-driven verification rules into LLM inference. We evaluated it on three BioRE benchmarks (SemRepGS, DDI, and ChemProt) using both proprietary and open-weight LLMs. To assess generalizability beyond benchmark datasets while reducing potential evaluation bias from LLM pretraining contamination, we conducted a temporal evaluation using 100 biomedical articles published in 2026. With the proprietary backbone, ANCHOR-RE outperformed direct LLM prompting, improving micro-F1 from 0.654 to 0.676 on SemRepGS, from 0.769 to 0.872 on DDI, and from 0.939 to 0.941 on ChemProt. On DDI and ChemProt, it also outperformed previously reported inference-only methods and approached fine-tuned or instruction-tuned systems without parameter updates. Similar performance gains observed with open-weight LLMs indicate that the benefits were not limited to the proprietary backbone. On the post-cutoff set, manual assessment of 500 randomly sampled predictions yielded a precision of 69%, maintaining consistent precision on previously unseen biomedical literature. Neuro-symbolic reasoning can improve the reliability of LLM-based BioRE without fine-tuning. Results across multiple benchmarks, model families, and post-cutoff literature support ANCHOR-RE as a practical training-free approach to biomedical literature mining.

RAG improves the factual grounding of LLM by incorporating external knowledge, but deploying RAG on mobile and edge devices remains challenging because retrieved context increases computation and memory. A direct way to reduce this cost is to retain only one retrieved chunk before generation, but the top-ranked retrieved chunk is not always the most evidence-supporting one, since retrieval similarity does not necessarily imply evidential sufficiency. Existing context-reduction methods can improve context quality, but often require additional LLMs or compressors that are costly under a strict mobile budget. In this paper, we study lightweight RAG chunk selection as an evidence-alignment problem. Our selector combines three complementary feature sources: question hidden states that represent LLM-side query intent, MoE routing-derived expert signals that capture the generator's internal routing structure, and retrieved chunk embeddings that preserve candidate-side evidence geometry. A compact multilayer perceptron maps these features to an evidence prototype in the chunk embedding space, and the candidate most aligned with this prototype is selected by cosine similarity. For stricter deployment budgets, we further introduce an optional task-aware feature selection strategy to reduce the selector input dimension. To support supervised evaluation, we construct semantic chunk-correctness labels based on evidence sufficiency rather than answer-string containment. Experiments show that the proposed selector consistently improves rank-1 evidence selection over mobile-applicable baselines by an average of 2.5%. These results suggest that using LLM-side query representations and MoE routing information and aligning them with retrieval-side candidate embedding is an effective and parameter-efficient strategy for mobile-applicable RAG chunk selection.

Vision-and-Language Navigation (VLN) requires an agent to follow a route-level instruction by executing its constituent steps from egocentric visual observations. Existing VLM-based navigators typically supervise both capabilities through next-action prediction alone, making progress-tracking errors difficult to distinguish from execution errors. When an agent deviates from the route, a corrective action label may recover the next movement but does not indicate whether the agent selected the wrong sub-instruction or failed to execute the correct one. Consequently, the agent may continue making decisions from an erroneous progress state. To resolve this ambiguity, we propose \textbf{Route2Step}, a framework that decouples semantic progress tracking from action generation through an explicit step-level interface. The Instruction Analysis Module ($\mathcal{M}_{\mathrm{IA}}$) predicts this state from the global instruction and visual history. Conditioned on the predicted state and recent observations, the Action Generation Module ($\mathcal{M}_{\mathrm{AG}}$) generates local action chunks. To supervise the progress state without manual temporal labels, E-SPA, a step-alignment procedure, associates sub-instructions with their corresponding portions of route-level demonstrations. These alignments enable state supervision for incorrect progress estimates, while direct action supervision is reserved for rollout groups that repeatedly fail under the correct active sub-instruction. On R2R-CE, Route2Step improves SR from 48.1\% to 55.3\% and SPL from 43.3\% to 48.2\%, using 190K state-level corrective samples while requiring only 11.5K directly action-supervised states. Experiments in real-world indoor and outdoor environments further demonstrate the practical applicability of Route2Step. The project page is: https://sisyphus-hxy.github.io/Route2Step/.

Generating a rigorous paper introduction with large language models (LLMs) remains challenging, since it requires coordinating background, gap identification, method and contribution within a coherent narrative. Existing solutions externalize this process as multi-stage prompts or agent workflows which are expensive and vulnerable to cross-stage drift. We propose StructPO, a struct-aware policy learning framework that internalizes the entire multi-stage writing workflow into a single-pass policy controlled by explicit stage tokens. StructPO introduces struct-aware credit assignment to decouple local stage quality from global coherence and refinement-guided optimization to internalize revision behavior into the first-pass policy. Experiments show that StructPO improves semantic alignment, structural rationality and inference efficiency over workflow-based baselines, generalizes to out-of-domain settings, and remains competitive with GPT-5.1 in human evaluation when scaled to Qwen3-32B. These results show that internalizing academic writing workflows through fine-grained policy optimization offers a viable alternative to costly external orchestration.

Long-term memory enables persistent personalization in LLM agents, but repeated memory-conditioned responses can cumulatively reveal protected attributes even when they are never stated explicitly. We formalize this threat as adaptive transcript privacy and introduce DP-MemView, a differentially private interface that privately selects public response-conditioning views and exposes those views---rather than raw memory---to the response LLM. Each private selection is charged to every protected attribute whose memory group intersects the read set. Per-attribute ledgers block any selection that would exceed its cap and return a fixed generic view instead. Under an explicit interface contract, we prove pure B_a-DP for the entire adaptive transcript. We also extend the result to stores that differ across multiple protected groups and bound how much observing the transcript can change an adversary's prior odds. We evaluate the online and preallocated modes with three response LLMs on a controlled adjacent-store benchmark and a public-corpus transfer track. Both modes keep transcript distinguishability near chance while preserving target-required personalization and overall response quality. Further diagnostics show that removing key safeguards causes mismatched output support, missing ledger charges, revealing side channels, or growing long-horizon leakage.

This innovative practice full paper describes an AI-powered knowledge graph platform that connects SQL errors to conceptual gaps in undergraduate and graduate database systems courses. Students learning Structured Query Language (SQL) frequently struggle with semantic errors that reflect conceptual misunderstandings rather than syntax mistakes. A query may execute yet return incorrect results due to gaps spanning related concepts; misusing NATURAL JOIN in place of an explicit subquery reflects intertwined misunderstandings of JOIN, GROUP BY, and HAVING. Autograding systems detect correctness but provide surface-level feedback without connecting errors to the conceptual structure of the course. Educational knowledge graph research has shown the value of structured concept representations for curriculum analysis and adaptive learning, but these approaches have not been applied to diagnosing SQL misconceptions from student submissions. We present a platform that automatically extracts course concepts and relations from instructional materials, links them to student submission traces through a graph database, and classifies errors at the concept level. We evaluate the platform across two database systems courses at two universities, one using real student submissions and one using simulated submissions, through an expert study with five participants and an automated evaluation using an LLM as a judge. Results show that 95.7% of extracted nodes were rated as at least somewhat valid and 63.8% of triplets were rated fully correct. Expert feedback confirmed that the generated graphs align with instructor mental models and that mapping errors to course concepts provides actionable diagnostic insight; evaluating impact on student learning remains future work.

Motion retargeting approaches often prioritize kinematic similarity over whole-body dynamics, contact consistency, and actuation limits, yielding references that are difficult for reinforcement learning (RL) policies to reproduce, particularly for contact-rich behaviors. We present a contact-implicit, direct simulation-based multiple shooting (DSMS) framework that transforms kinematically feasible references into dynamically feasible whole-body trajectories. By embedding a differentiable simulator within a nonlinear program, DSMS resolves contact, friction, impacts, self-collision, and joint limits internally while enforcing tracking, actuation, and task constraints without prescribing a contact schedule or introducing explicit contact constraints. Compared with existing retargeting methods, DSMS accelerates motion-imitation RL training and yields policies with high success rates and low tracking error. We further demonstrate zero-shot sim-to-real transfer on the Unitree G1 through command-conditioned contact-rich crawling and a highly dynamic 180-degree jump-turn.

Diffusion policies have shown strong performance in learning complex, multi-modal behaviors for robotic manipulation. However, their application to contact-rich disassembly tasks remains limited by a key trade-off: the iterative denoising process introduces inference latencies that makes high frequency control difficult, which is essential for realizing dynamic interactions such as chiseling and prying. Recent action-chunking techniques mitigate latency but use an open-loop execution window, rendering the system blind to rapid force transients caused by fracture events. To bridge this gap, we introduce the Diffusion Policy Augmented by Fast Trajectory Generation (DPA-FTG). Compared to recent visual-tactile approaches that focus on positional correction, DPA-FTG decouples low-frequency planning from high-frequency force regulation. At the high level ($5$ Hz), a conditional diffusion model predicts a sequence of latent parameters for selecting a strategy from a learned vocabulary of task primitives. At the low level ($60$ Hz), a lightweight, force-conditioned policy acts as a neural impedance controller, modulating execution in real-time to maintain contact stability. We validate our approach on a bimanual battery disassembly task involving the separation of a compliant sheet. Experimental evaluation demonstrates that DPA-FTG outperforms state-of-the-art baselines, including Reactive Diffusion Policy (RDP).

Foundation models place language throughout embodied agents, but its presence does not show what it contributes or how well that contribution is grounded. This survey separates these two questions. We define five non-exclusive functional roles for language: Specification, Embodied Representation, Action Orchestration, Grounding Regulation, and Execution Coupling. For each role, we trace the path from linguistic content to its embodied consumer and identify the observations or interventions that can test the claimed responsibility. Applying this framework to the reviewed literature reveals a recurring gap between functional use and evidential support. Interpretable or revised linguistic intermediates may be incorrect, go unused, or fail to affect later behavior. Even when actions are directly conditioned on language, system-level success does not by itself isolate language's contribution. We therefore evaluate grounding claim by claim, asking whether the reported evidence supports the specific responsibility assigned to language. Using role claims rather than architectures as the unit of comparison allows us to compare modular and end-to-end embodied agents without extending conclusions beyond the reported evidence.

We present VIVID (Vietnamese Idioms for Validation and Interpretation Depth), the first systematic benchmark for evaluating culturally grounded figurative language understanding in Vietnamese. VIVID comprises 1,636 idioms and proverbs annotated with five complexity traits (literal expressions, pragmatic nuances, Sino-Vietnamese terms, uncommon vocabulary, folk knowledge) and seven semantic themes. We establish an evaluation framework combining generative and discriminative tasks, proposing an LLM-as-a-Judge approach with aspect-based prompting validated against human judgment (Cohen's kappa = 0.792). Evaluating eight state-of-the-art models reveals critical gaps: Vietnamese-specialized models drastically underperform multilingual systems (VinaLLaMA-7B: 0.13 vs. GPT-4o: 2.46), and even top models achieve less than 50% of maximum scores. Notably, few-shot prompting does not universally improve performance, with GPT-4o exhibiting degradation due to stylistic overfitting. Our analysis exposes systematic failures including literal over-interpretation, lexical gaps, and pragmatic flattening, demonstrating that current models lack cultural competence for nuanced figurative interpretation. VIVID provides an essential tool for advancing figurative language understanding in culturally rich contexts.

Video large language models (VideoLLMs) achieve strong video understanding performance, but their inference remains expensive due to the large number of redundant spatio-temporal visual tokens in long videos. Existing token pruning methods alleviate this cost by reducing redundant tokens, yet most of them rely on segment-level local pruning, where videos are partitioned into isolated segments and tokens are selected independently within each segment. Such designs may under-preserve short but semantically dense segments and discard tokens that appear non-salient locally but remain critical from a global perspective. To address this issue, we propose GSTEP (Global Spatio-Temporal Density Pruning), a plug-and-play pruning framework that models video as a continuous spatio-temporal information flow. GSTEP constructs a token-level spatio-temporal density by combining a continuous temporal density, obtained from a smoothed centered frame-level change signal, with intra-frame spatial density, and then performs global token sampling by jointly balancing information density and coverage. Extensive experiments on multiple VideoLLMs and public benchmarks demonstrate that GSTEP consistently achieves strong accuracy-efficiency trade-offs and generalizes well across model architectures and evaluation settings. On LLaVA-OneVision-7B, GSTEP prunes 75% of visual tokens, preserves up to 100.2% of the original average performance across benchmarks, and achieves a 1.17 end-to-end speedup.

Frontier AI model developers increasingly rely on layered safeguards to prevent catastrophic misuse, but little public evidence exists on how much protection these safeguards provide, or how consistently across developers. We introduce the FAR.AI Minimal Standard for Safeguards, Version 1.0: a taxonomy of 67 readily accessible static jailbreak techniques, a method for composing them into a very large attack space, and a benchmark of flagship models against a sample of it. We evaluate Claude Fable 5, GPT-5.6 Sol, Gemini 3.1 Pro, and Grok 4.5 on two complementary datasets totalling 360 attacker goals spanning chemical, biological, radiological/nuclear and explosive (CBRNE) threats and offensive cyber, using a three-stage funnel to identify universal jailbreaks: single prompt templates that elicit operationally compliant responses on over 75% of a domain's goals. We also introduce a cost-to-jailbreak metric that models attacker spend directly, with right-censored lower bounds where no universal jailbreak was found. Robustness is highly uneven: the cost to break these models varies over a hundredfold. Random search over our technique pool found 63 universal jailbreaks against Grok 4.5 and 18 against Gemini 3.1 Pro, at an average cost of roughly $58 and $278 per jailbreak found; expert-guided composition raised these to 385 and 231. Neither Claude Fable 5 nor GPT-5.6 Sol yielded any universal jailbreak under either strategy. Because meeting the Minimal Standard requires only defenses already publicly described and deployed in production elsewhere, these gaps appear closable with current techniques. We recommend defense-in-depth combining reasoning, activation, and input/output monitoring. Results are maintained at leaderboard.far.ai.

Reinforcement learning (RL) has emerged as an effective method for enhancing the reasoning capabilities of large language models (LLMs). However, existing methods suffer from insufficient precision in feedback on generated answer trajectories and exhibit the phenomenon of problem difficulty drift. To address these challenges, we propose CVPO - Curriculum-guided Value-Variance Policy Optimization. At the response trajectory level, we find that token-level value-variance correlates with exploration intensity. Our theoretical analysis shows this variance bounds policy update magnitude. We then use the estimated trajectory value-variance to quantify the intrinsic randomness in generation. Based on this, we design a variance-aware advantage adjustment mechanism for different reward types. At the question level, we introduce a dynamic curriculum weighting method that adapts to question difficulty. This helps the model focus on tasks matched to its current ability during each training stage. Experimental results show our method outperforms strong value-based baselines like VAPO. It achieves better performance and stronger exploration, enabling more accurate and robust reasoning in language models across various math tasks.

Merchant risk control at large payment platforms screens tens of millions of merchants daily, where false positives harm legitimate merchants and false negatives leave harmful activity undetected. The hardest cases require jointly understanding a merchant's textual profile and long behavioral sequence. Large language models (LLMs) excel at text but cannot natively model such sequences, while adapting them often causes catastrophic forgetting. We present SeqLLM, a framework that adds behavioral-sequence modeling to a pretrained LLM while preserving its language ability. SeqLLM combines three components: a compact discrete vocabulary that represents behavioral events as native tokens; a lightweight projector, trained with a two-stage alignment curriculum, that grounds these tokens in the LLM's semantic space; and prefix-guided capability injection, which acquires sequence-modeling ability through task-prefixed supervised fine-tuning rather than continual pre-training. SeqLLM is deployed at WeChat Pay, screening millions of merchants daily. Against the production DeepSeek-based LLM baseline, it raises screening precision from 92.0% to 97.5%. Its pretrained behavior-token embeddings also improve Precision@Top-0.01% by 26.8 percentage points in a production fraud detector serving billion-scale transaction traffic. Beyond payments, SeqLLM achieves state-of-the-art results on public recommendation benchmarks. On MovieLens and Amazon, it surpasses the strong User-LLM baseline by up to 32% relative Recall@5 while retaining markedly stronger language ability. On RecIF, it improves Pass@32 by 14.2% over the full OneRec-8B pipeline using only one-fifth of its GPU-days.

Recent Vision-Language-Action (VLA) models almost universally take robot proprioceptive state as input, yet wire it in incompatible ways -- serialized into text prompts, projected into the vision-language prefix, or fed directly to the action expert -- and almost always as a single current frame. Three questions remain open: (1) whether, and on which tasks, current state actually improves closed-loop control; (2) how much state history helps, and whether its benefit reflects genuine temporal variation rather than added conditioning capacity; and (3) where state should enter the model -- the vision-language backbone or the action-generation module. We answer these questions through controlled experiments on a flow-matching VLA, fixing the backbone, training data, action representation, and evaluation protocol throughout. We implement five representative interfaces -- discrete state prompt, VLM prefix, action prefix, state expert, and feature modulation -- under matched implementation details, and evaluate them on 45 atomic tasks spanning three task families plus 20 composite tasks; we then sweep the state-history length from 1 to 96 frames to examine how historical state information affects model performance. The experiments yield systematic answers to all three questions, distilled into testable design principles for state-aware VLAs.

Model Predictive Control (MPC) delivers constraint-aware control, but its reliance on online optimization limits its use on systems with fast dynamics, high-dimensional models, or long horizons. Existing GPU implementations typically treat the device as a linear-algebra accelerator, leaving the optimization loop dependent on repeated kernel launches and high-latency memory transfers. This paper introduces CUDA MPC, a GPU-native MPC framework that co-designs the optimization algorithm, execution model, and memory architecture for CUDA hardware. CUDA MPC pairs a parallel-in-horizon alternating direction method of multipliers (ADMM) splitting with a fused CUDA kernel that runs the entire iterative solve on the device. Intermediate optimization variables stay in low-latency, on-chip shared memory, and a localized atomic-flag protocol synchronizes only adjacent horizon blocks, minimizing host intervention, kernel-dispatch overhead, and global-memory traffic. Across six nonlinear robotics benchmarks spanning increasing state dimension and constraint density, CUDA MPC sustains real-time rates at horizons one to two orders of magnitude longer than CPU solvers: it solves an optimization-based collision-avoidance parking problem with 100 s of lookahead within a 0.1 s sampling interval, and is the only solver evaluated that achieves both real-time execution and collision-free coordination for a centralized 10-agent swarm, where acados and CasADi return no feasible solution and require 3.5 s and 4.5 s per solve. Against tensor-framework implementations of the same ADMM splitting, the fused kernel is up to $965\times$ faster.

Long-context reasoning remains a critical bottleneck for large language models, as recent recurrent-memory approaches face two inherent challenges: sequential chunk-wise updates can overwrite early critical evidence with later irrelevant content, and serial inter-chunk dependencies limit parallelism and cause latency to increase with context length. To address these issues, we propose PI-Mem (Parallel-Iterative Memory), a mechanism that processes all chunks in parallel and iteratively refines a shared memory over a bounded number of turns. In each turn, PI-Mem reads all chunks in parallel conditioned on the current memory, selects new or complementary evidence from each chunk, and merges the selected evidence into a compact shared memory for the next turn. To discourage redundant turns, we optimize the workflow through reinforcement learning with an auxiliary turn-efficiency reward, enabling the model to adaptively exit once sufficient evidence has been accumulated. We evaluate PI-Mem with Qwen3.5-35B-A3B and Qwen2.5-7B on the HotpotQA benchmark across context lengths up to 3.6 million tokens and find that it outperforms the recurrent-memory baseline by +6.25 and +7.81 absolute points while achieving 6.1$\times$ and 2.1$\times$ inference speedups, respectively. These results demonstrate that PI-Mem breaks the accuracy--efficiency trade-off in long-context reasoning and provides a scalable approach to complex multi-hop question answering over extremely long documents.

Reasoning-enhanced large language models have achieved remarkable improvements in planning tasks, yet their deployment in embodied systems remains impractical due to prohibitive inference delays-often exceeding minutes per planning instance. The fundamental bottleneck stems from the serial nature of existing paradigms: models must complete all reasoning before any action execution, leaving execution time windows entirely unexploited. We introduce PACE (Planning with Adaptive Cognitive Effort), a framework that enables interleaved reasoning and execution through two key innovations: an Interleaved Think-Act architecture that pipelines cognitive processing with action execution, and a Dynamic Budget Allocator that adapts reasoning token budgets to available execution time windows. On the Robotouille benchmark using Qwen3-8B-AWQ, PACE achieves a 10% success rate-representing a 67% improvement over the ReAct+Think baseline-while delivering 6.9 times acceleration in thinking time compared to unconstrained reasoning. The framework hides 66.8% of thinking time within execution windows, demonstrating that strategic cognitive effort allocation can simultaneously improve both planning quality and time efficiency. These results provide evidence that time-aware architectural innovations enable reasoning models to operate in latency-sensitive embodied domains where they were previously impractical.

This paper addresses the challenge of providing kinesthetic feedback in bilateral teleoperation by designing a wearable, lightweight (20 g), and compact haptic device, the HJ-Haptic, utilizing a honeycomb jamming mechanism for object stiffness rendering. The HJ-Haptic device can vary its stiffness, from 1.15 N/mm to 2.64 N/mm, using a 30 kPa vacuum pressure. We demonstrate its implementation in a teleoperation framework, enabling operators to adjust grip force based on a reliable haptic feedback on object stiffness. A three-point flexural test on the honeycomb jamming mechanism and teleoperated object-grasping tasks were conducted to evaluate the device's functionality. Our experiments demonstrated a small RMSE and strong correlations in teleoperated motion, stiffness rendering, and interaction force feedback. The HJ-Haptic effectively adjusts its stiffness in response to real-time gripper feedback, mimicking the sensation of direct object grasping with hands. The device's use of vacuum pressure ensures operator safety by preventing dangerous outcomes in case of gas leakage or material failure. Incorporating the HJ-Haptic into the teleoperation framework provided the reliable perception of object stiffness and stable teleoperation. This study highlights the potential of the honeycomb jamming mechanism for enhancing haptic feedback in various applications, including teleoperation scenarios, as well as interactions with extended-reality environments.

Latent diffusion models (LDMs) have recently significantly advanced embodied learning in constructing powerful embodied manipulation world models. However, despite the remarkable performance, existing LDMs predominantly rely on Variational Autoencoders (VAEs) optimized for natural scenes while failing to account for the unique characteristics of embodied manipulation scenarios, yielding latent representations that are neither compact nor controllable, thereby hindering efficient training of LDMs and precise robotic control. To solve this problem, we present EmbodiedVAE, a novel video VAE that provides compact yet controllable latent representations tailored for the robotic manipulation world models. Specifically, EmbodiedVAE adopts a dual-encoder, single-decoder architecture with an asymmetric spatio-temporal compression module, which automatically disentangles the robot arm's motion from background environment, resulting in overall compactness while providing explicit embodied latent to support fine-grained action control. To further preserve the temporal consistency of learned robotic motion latent, we introduce an optimal-transport-based consistency module that explicitly enforces motion fidelity and inter-frame coherence. Extensive experiments demonstrate that our proposed EmbodiedVAE achieves superior reconstruction quality with high compression rate, while enabling more precise action control in robotic manipulation scenarios with an average of 2dB PSNR improvement over state-of-the-art video VAEs.

Speculative decoding verifies a tree of draft tokens in one target-model forward pass. For a mixture-of-experts (MoE) target, however, parallel verification can activate the union of the experts selected by all tree nodes, even though only a small subset of those nodes reaches the accepted output. Token count, activated-expert union size, and expert-weight traffic are therefore distinct cost measures: reducing the token workload need not shrink the expert union proportionally, and under offloading, transfer traffic also depends on cache residency. We introduce AcceptMoE, a verifier-side expert selector that combines target-router scores with offline-estimated commitment probabilities and automatically adjusts the number of eligible experts for each verification block, eliminating the need for a user-specified expert budget. Under offloading, AcceptMoE conditions expert eligibility on cache residency instead of predicting natural routes and prefetching the corresponding expert weights. Although constraining target-expert eligibility changes the model distribution, across 12 model-task pairs spanning three MoE targets and four benchmarks, AcceptMoE's mean accuracy is 0.27 percentage points lower than that of EAGLE-3 speculative decoding with natural routing. Served with SGLang at batch size one, it reaches 1.290 times the throughput of this baseline with all expert weights in GPU memory, and 2.06 times under physical expert offloading, while reducing host-to-device traffic by 73.6 percent to 77.1 percent.

Large language models (LLMs) have demonstrated impressive performance in MQM-based translation quality (TQ) evaluation, and recent advances in large reasoning models (LRMs) promise even greater improvements. However, both LLMs and LRMs are computationally expensive to deploy at scale, while small language models (SLMs)---though much more efficient---struggle with the complex reasoning required for evaluation tasks. In this work, we present an extensive empirical study benchmarking SLMs, LLMs, and LRMs across a wide range of TQ evaluation setups, providing a comprehensive view of the current landscape and establishing best practices. To address the scalability challenge, we introduce TQLite, a novel distillation framework that enables SLMs to approach the MQM evaluation performance of the best LRM-based evaluators. Our approach leverages a multi-LRM jury to generate high-quality synthetic training data via practical data curation techniques and aggregation of evaluation responses across a diverse panel of models. Our results demonstrate that SLMs trained via TQLite achieve strong MQM evaluation performance that far exceeds off-the-shelf evaluation capabilities of standard SLMs, offering a scalable and cost-effective alternative to LLM- and LRM-based evaluators.

Most multiple-choice question (MCQ) benchmarks evaluate Large Language Models (LLMs) only by whether they select the correct answers. This binary scoring treats all incorrect responses alike, even though an LLM's preferences among incorrect options may contain systematic and useful information about its behavior and ability. We introduce the LLM Nominal Response Model (LLM-NRM), an option-aware psychometric framework that models the full distribution over answer choices to jointly estimate LLM ability and option-level item characteristics, while separating model-specific response calibration sharpness, positional preference, and difficulty-dependent fallback behavior. Across 189 LLMs and 31,554 items from 14 benchmarks, LLM-NRM predicts held-out LLM-item interactions more accurately than binary Item Response models and conventional nominal-response baselines, and its ability estimates achieve the strongest Spearman correlation of 0.920 with the external human-preference Arena.ai Elo leaderboard. Distractor identity contributes +101% additional Fisher Information per item beyond correctness, and incorrect responses alone recover full-information ability estimates with Spearman 0.943. The learned item parameters also enable efficient benchmarking, where 41 selected items preserve the full-bank ranking with Kendall's correlation 0.85, corresponding to a 770 times reduction. In conclusion, we show that incorrect answers carry distinct and useful measurement information rather than representing equivalent mistakes.

Vision-Language-Action (VLA) policies trained by behavior cloning fail silently: from the action stream alone, a collapsing rollout looks much like one making clean progress, because imitation supplies no notion of progress. Reinforcement learning would supply one, but it is impractical here, where real-robot experience is costly and deformable food resists simulation. The cheap alternative, a terminal success / failure bit, is learnable in principle yet far too sparse to say when a rollout went wrong. We argue that the per-frame label, not the architecture, is the hard part: to be useful it must be dense, continuous, and correctly shaped. We present ValueFormer, a compact policy-agnostic causal transformer over a frozen DINOv3 backbone that emits two per-frame signals in one forward pass: a smooth Monte Carlo value, V_mc, for advantage estimation and a sharp binary value for online mistake detection, targets that pull in opposite directions by design. Failed episodes are labeled with a stage-aware, success-then-decay return that preserves the success curve before the failure stage, and detection is supervised from mistake intervals rather than a single failure time, so mistakes the policy recovers from also carry signal. On a real-robot bimanual sandwich-assembly task 1,427 episodes), a critic-derived per-frame training weight lifts task completion from 70% to 85% (within noise at n=20), and a batched bf16 encoder cuts the live serving cost 3~5 times so the critic runs at 2 Hz alongside the policy on a single GPU.

Industrial humanoid robots are constrained less by locomotion or manipulation capability than by the immaturity of functional safety certification for legged platforms. The root difficulty is that the safe state of a legged robot is an actively-controlled state, which violates the fail-passive assumption underlying ISO~13849-1 / EN~60204-1: removing power from a walking biped causes an uncontrolled fall, so classical de-energization is itself a hazard. We term this the fail-passive gap and use a certified external safety chain (light curtain, emergency stop, fail-safe input, fail-safe PLC, and wireless PROFIsafe) as an instrument to locate it precisely: because the external chain is closed and quantifiable with established methods (PFHD, DC, CCF, PL/SILCL), the residual uncertifiable element is pinpointed to the robot-side reaction chain. Using a Siemens fail-safe S7-1500 emergency-stop reference, we show its certifiable Reaction subsystem is contactor-based power removal (Stop Category~0)---exactly the element a balancing humanoid cannot have. We deliberately do not claim end-to-end certified PL~e / SIL~3. We validate the approach on a Unitree G1 EDU pick-and-place cell in a 3m x 1.5m semi-enclosed workspace, and contribute a humanoid-specific analysis of the active safe state (fall-as-hazard, single-support stop bounds, balancing-policy residual risk, ISO~13855 separation) and a provenance-labeled timing budget. Hosting an industrial software-defined automation (SDA) controller on the robot, co-located with the balancing policy, moves robot-side PROFINET/PROFIsafe reception onto a standardized IEC~61131-3 interface; because the G1's onboard compute is not safety-rated hardware, this endpoint is not a certified safety runtime, which reinforces rather than resolves the fail-passive gap and localizes it to the SDA-to-balancing-policy interface.

In robot teleoperation, haptic feedback can be used to help human operators accomplish dexterous manipulation tasks. However, existing haptic feedback methods try to replicate high-fidelity sensory haptics that are felt in real world interactions, which are constrained by the sensing and feedback hardware capability and may lead to higher workload. To addresses these limitations, this work introduces semantic haptics for teleoperation, which uses abstract haptic patterns to convey critical information about robot states. We categorize robot states into "Confirmations" and "Exceptions", implement a modular haptic rendering pipeline in robot simulation, and deliver semantic haptic feedback to operators through pneumatic and vibrotactile wristbands. This simplifies hardware requirements and enables one-to-many mappings between haptic patterns and robot states. Through three evaluation studies, we identify the most effective semantic haptic design for a common pick and place teleoperation task and compare semantic haptics to other teleoperation feedback approaches including sensory haptics and visual feedback. Results suggest that while semantic haptics performs similarly as other feedback in unimanual tasks, it achieves superior performance in bimanual tasks, with reduced task workload, increased situational awareness, and overall preference.

Continually improving agents require dynamic interaction feedback beyond static supervision, yet direct real-environment interaction is costly, slow, unsafe, and hard to parallelize. World modeling offers a natural intermediate proxy that allows agents to query lower-cost, more controllable feedback before committing to real actions. Classical world models instantiate this proxy primarily through future physical-state prediction, a formulation useful yet narrow for agents that require actionable feedback beyond raw state transitions. In this work, we conceptualize Agent-Centric Interactive World Proxies, shifting the fundamental paradigm from physical state transitions to agent-usable information transitions, such as execution outcomes, retrieved experiences or skills, and verification signals, broadening the scope of world modeling to provide versatile feedback for continually improving agents. To systematically map this design space, we organize world proxies into six functional forms based on their feedback modalities: dynamics, spatial, execution, memory/experience, skill, and reward/verification proxies, which together characterize the primary ways world modeling serves agent improvement. We further analyze how these proxies empower agents across three progressive levels: L.1 Inference-Time Guidance, where proxy outputs enrich in-context information for superior decisions; L.2 Training-Time Optimization, where proxy outputs yield rewards, critiques, or synthetic rollouts for policy learning; and L.3 Agent-Proxy Co-Evolution, where real-environment evidence continuously updates both the proxy and the agent for co-evolution. Ultimately, this work recasts world modeling into an agent-centric paradigm, establishing a roadmap for building world proxies that empower agents to plan better, learn faster, and evolve continually.

Learning generalizable robot manipulation policies requires large-scale and diverse demonstration data. Egocentric human manipulation videos offer rich scene and task diversity, and prior work has shown that retargeting and rendering such videos into robot-format data can yield effective per-task policies at small scale. However, whether this approach can provide pretraining benefits for vision-language-action models at scale remains unexplored. We present \textbf{Ego2Robot}, a scalable pipeline that converts egocentric human manipulation videos into robot training data through action retargeting, robot-arm visual synthesis, and multi-level quality curation. Ego2Robot supports both curated datasets and in-the-wild videos, producing 18,561 hours of robot training data spanning 15 robot morphologies, making it the largest ego-to-robot dataset to date. To evaluate generalization, we extend RoboTwin2.0 with disentangled perturbation axes covering visual appearance, scene layout, embodiment morphology, and task semantics. Experiments show that joint pretraining on Ego2Robot-synthesized and robot data consistently improves out-of-distribution generalization across multiple perturbation types, with benefits validated on real-robot deployment. Project page: https://www-ye.github.io/ego2robot_blog/

World Action Models (WAMs) augment robot policies with action-conditioned predicted futures, but a plausible future alone does not justify changing the action that a bimanual policy would execute. We present CoWAM, a selective intervention layer that expresses synchronization, role compatibility, and collision convergence as coordination contracts. Each contract combines typed admissibility checks with event-conditioned verification and calibrated intervention gates. CoWAM preserves the nominal action unless an alternative satisfies every active obligation and provides a clear, low-risk improvement; when the nominal action is also inadmissible, it invokes a predefined abstention fallback. To separate selector quality from proposal quality, all methods operate on identical candidate pools and commit their decisions before shared oracle labeling. Across eight simulated bimanual tasks, CoWAM improves coordination-valid selection by 16.7 percentage points over the contract-only variant and raises closed-loop success by 9.6 percentage points over the strongest selective baseline, while keeping harmful interventions below 1%. Together, these results establish coordination contracts as an effective interface for conservative policy intervention with predicted world-action evidence across coordination-rich bimanual tasks.

Quadruped robots are a promising platform for search and rescue missions because they can navigate cluttered indoor environments that may be restrictive for wheeled systems. However, in unknown rescue scenarios, autonomous exploration must balance map expansion with the likelihood of finding victims, which is not explicitly addressed by clas- sical frontier selection strategies. This paper presents a situation aware frontier prioritization method for single robot quadruped search and rescue. The proposed approach preserves the frontier exploration framework, but extends frontier ranking with information gain, observation deficit, rescue relevance, terrain penalty, and travel cost. The method is eval- uated in Gazebo simulation with a quadruped robot in two indoor rescue scenarios with different levels of difficulty. The first scenario is used as a sanity check, while the second introduces stronger clutter and frontier ambiguity. Experimental results show that all methods perform reliably in a simple scenario, whereas in a complex scenario is different. In that setting, the proposed method achieves the highest completion rate and the highest victim recovery among the evaluated approaches. These results indicate that situation aware frontier prioritization is beneficial when frontier choice becomes nontrivial and rescue utility must be balanced against generic exploration objectives.

Action chunking---predicting and executing multiple actions instead of a single action---has proven to be a critical component for learning effective robotic control policies. However, our precise understanding of why action chunking improves performance has remained limited. In this work we seek to close this gap. Through rigorous experimental evaluations in both simulated and real-world settings, we show that existing hypotheses for the success of action chunking---temporal consistency, horizon reduction, and representation learning---fail to explain the success of action chunking. Instead, we find that action chunking benefits from greater non-Markovian expressivity and reduced compounding error compared to Markovian policies, but, in many settings of interest, these effects can be fully captured by delayed policies, which at each step predict a single action based on the observation $k$ steps in the past. We then show that there exists an additional benefit of action chunking that we refer to as implicit ensembling. In particular, by learning a diversity of temporal relationships (that is, $a_t | o_t, a_t | o_{t-1}, \ldots$), action-chunked policies exhibit behavior matching that of a model ensemble, increasing their robustness and generalization ability over policies that only learn a single temporal relationship. Building on these insights, we show that in simulated and real-world robotic control settings, we can match the performance of action chunking without action chunking---by deploying an action chunking policy as an ensemble of policies with randomized delays. Furthermore, we propose a policy class that amplifies the benefits of action chunking by explicitly instantiating an ensemble, and which we show significantly improves over the performance of action chunking in many domains.

Reachability analysis for visuomotor policies is difficult because large visual encoders make end-to-end set propagation computationally expensive and excessively conservative. We therefore freeze the visual encoder and confine set propagation to a low-dimensional interface between it and the downstream policy, with the interface set calibrated from held-out camera-pose perturbations. Propagating this set through the policy with zonotopes yields a terminal output-enclosure width that set-based training optimizes directly. During evaluation, camera-pose perturbations are sampled from the prescribed distribution, and rollout-level split conformal calibration converts the resulting action-deviation scores into a probabilistic reachable-action radius with finite-sample coverage. In controlled manipulation experiments, set-based training reduces this radius while preserving closed-loop task capability, and matched behavior-only, observational-consistency, and pointwise-adversarial controls all leave a larger radius.

Vision-Language-Action (VLA) models excel in robotic manipulation but suffer catastrophic performance drops when canonical instructions are simply paraphrased. Although this brittleness is typically addressed through costly data scaling, our probing reveals that the root cause is architectural rather than a lack of semantic understanding. Specifically, we demonstrate that current VLAs successfully retain the correct task identity internally. The failure actually stems from the joint encoding of dynamic visual observations and text, which introduces systematic feature shifts. Because the downstream action policy is highly vulnerable to these variations, it fails to translate the preserved semantics into correct control commands. To resolve this structural bottleneck, we propose Grounded Semantic Re-binding (GSR), an elegant intervention that bypasses unstable joint routing by explicitly fusing independently extracted task semantics with native visual features to train a completely re-initialized action expert from scratch. This targeted intervention dramatically restores paraphrastic invariance using only canonical demonstrations. On the LIBERO-Para benchmark, GSR improves success rates by up to 44.6 percent. It enables lightweight models to rival massively scaled baselines and pushes state-of-the-art models to a new record PRIDE score of 70.4, outperforming the recently introduced large-scale pretrained model Xiaomi-Robotics-0 in instruction generation capabilities. Building on these insights, we also introduce ParaVLA, a natively decoupled 0.33B-parameter model exhibiting near-perfect robustness to instruction rewording. Ultimately, our work proves that robust semantic grounding can be achieved through elegant structural design, bypassing the inefficient brute-force data scaling paradigm.

Sim-to-real policies are designed under nominal dynamics, but target-system trials may yield only a few isolated one-step transitions. We study pre-execution certification of a fixed control sequence, such as an action chunk produced by a learned policy. If the sequence reaches an unobserved state-input region, the observations remain consistent with target systems whose trajectories separate along it by an arbitrarily large amount. Any deterministic certifier sound for all of them must then decline to certify or return a reachable tube with arbitrarily large projected width. For bounded smooth classes of the target-nominal model error, we derive a finite plan-dependent projected-width lower bound. These results expose a trilemma among uniform trajectory containment, finite projected width, and unrestricted model-error behavior beyond the observations. ForeReach requires a supplied componentwise Lipschitz bound on the model error. Observed transition pairs can refute this declaration but cannot establish it outside the observed locations. Conditional on a valid declaration, our method constructs a set-membership envelope for the model error, propagates a zonotopic reachable tube, and certifies only when propagation remains within the certification domain and every projected tube slice avoids the unsafe set. In two benchmark systems, calibration baselines may remain narrow after losing trajectory containment outside data support, whereas our method declines to certify unsupported sequences and recovers certification when relevant target data and sufficient obstacle clearance are available.

Humanoid motion trackers perform reliably within learned tracking distributions, but falls can move the robot into low-height, contact-rich states from which an advancing command is temporarily unreachable. Tracking-only policies may chase infeasible references, producing rapid, large-amplitude limb corrections that increase risk to the robot and its surroundings. We present StableMimic, a unified tracker trained beyond the nominal tracking distribution. Perturbed resets around multiple human get-up references expose prone, supine, off-balance, and intermediate ground-contact states, shaping structured recovery that returns the robot to the trackable region. Because tracking and recovery occupy markedly different state--action distributions, StableMimic uses dedicated experts for each regime and a proprioceptive gate that continuously blends their actions. A hidden successor-state objective teaches human-reference-shaped recovery without exposing reference identity or phase to the deployed Actor; deployment requires no get-up reference, recovery command, trajectory retrieval, or external policy switch. On the complete retargeted LAFAN1 dance subset, StableMimic achieves the lowest errors on all four tracking metrics among five methods. Across 100 matched push-to-fall trials per method, it recovers in 100/100 and attains the lowest values on six of seven post-fall motion and load measures, supporting improved interaction safety under this protocol. Real Unitree G1 dance and standing-reference deployments qualitatively demonstrate bounded limb motion, autonomous recovery, and command resumption.

This paper studies the distributed online control problem over a network of linear time-invariant (LTI) systems in the presence of adversarial disturbances and time-varying convex costs. The network cost is characterized by the summation of local cost functions, where each local function is sequentially revealed only to the corresponding agent. The goal of each agent is to generate a control sequence, using only local observations and neighbor communication, that competes with the best {\it centralized} linear policy in hindsight. We extend the recently proposed Online Spectral Control framework from the centralized setting to the distributed setting. In particular, each agent applies a spectral controller obtained by convolving past disturbances with the leading eigenvectors of a Hankel matrix, while the controller parameters are updated through a distributed online gradient descent step over the local surrogate costs. We formulate this problem this problem as a {\it regret} minimization problem based on the spectral parameterization, and under standard assumptions, we establish a sublinear regret bound of $O(\frac{\sqrt{T}\text{poly}(\log T)}{γ^3})$, where $T$ is the time horizon and $γ$ denotes the stability margin. The resulting bound also captures the dependence on the network size and connectivity.

External inputs to local flexibility markets, network models, demand scenarios, flexibility offers, and asset thermal, economic parameters do not change the clearing methodology, yet they are routinely defined under heterogeneous assumptions, which hinders reproducible benchmarking. This paper provides a complete, methodology-agnostic input set built around a modified CIGRE MV network with prescribed base-load and stress-load conditions, synthetic flexibility offers, wholesale prices, and transformer and cable parameters. Baseline AC operating points (voltages, loadings, and aging) are reported for both days so that alternative flexibility-request and clearing methods can be applied to the same inputs and compared fairly. The associated data are released openly on Zenodo.

World Action Models (WAMs) couple robot action prediction with video world models. Existing WAMs with shared-backbone and Mixture-of-Transformers designs generally tie the depth of the action module to that of the video backbone, resulting in substantial computational overhead and high inference latency. To address this limitation, we introduce Dock of Transformer (DoT), a video-centric design principle that treats a pretrained video Transformer as a representation hub and connects lightweight output-heads through docking interfaces. This enables flexible output-head design while providing direct access to representations from all layers of the backbone. We then introduce \textbf{Faster-WAM}, an instantiation of DoT for WAMs, which docks a single-layer action head onto a 30-layer video backbone. The docking interface fuses keys and values from all video layers and applies RoPE realignment. Without additional embodied pretraining, Faster-WAM achieves competitive performance on LIBERO and RoboTwin 2.0 while demonstrating strong out-of-distribution generalization on LIBERO-Plus. Faster-WAM also achieves the lowest end-to-end latency in our controlled comparison, requiring only 66.5 ms per inference --- a \(3.2\times\) speedup over Fast-WAM. Overall, these results demonstrate that the video-centric DoT architecture supports flexible task-specific head design while delivering low inference latency, strong action-prediction performance, and robust generalization.

In offline reinforcement learning (RL), the distribution shift between behavioral data and the learned policy can lead to erroneous \emph{Q}-value estimation, thereby misguiding the direction of policy optimization. To address this issue, we develop a behavioral advantage corrected policy evaluation (BAC-PE) approach, which utilizes the \emph{Q}-function of the behavior policy to correct the learned policy's \emph{Q}-function, thus mitigating pessimistic conservatism and overestimation bias. Furthermore, the convergence of BAC-PE is analyzed theoretically, and an upper bound on the difference between the learned \emph{Q}-function and the true \emph{Q}-function is derived. To alleviate distribution shift, this work employs diffusion models to represent both the behavior policy and the learned policy, performing distribution matching for accurate policy regularization. Additionally, \emph{Q}-value guidance is incorporated into the training process to achieve effective policy improvement. By combining BAC-PE with diffusion policy modeling, we propose the diffusion policy with behavioral advantage correction (DPBAC) algorithm. Compared to existing offline methods, DPBAC demonstrates stronger policy representation capabilities and effectively mitigates the bias in \emph{Q}-value estimation. Experimental results on multiple domains of D4RL tasks show that DPBAC achieves superior performance, with notable advantages over state-of-the-art (SOTA) algorithms.

Humans perform long-horizon manipulation by retaining knowledge of what earlier actions have established while continuously adapting the motion underway. By contrast, action-chunked vision-language-action (VLA) policies repeatedly replan from the current input at each query. Existing methods preserve either long-term task evidence through memory or short-term motion through action reuse and ensembling, leaving the cross-query handoff incomplete. We introduce ChainVLA, a 1.2B-parameter VLA policy that chains successive queries through a joint and revisable execution state. Progress Context combines a recurrent Working State with sparse event memory to carry observation-derived task progress, while Motion Tail feeds the preceding prediction's unexecuted continuation into state construction and action generation. Together, the two components condition a decoder that regenerates each action horizon under the latest observation, allowing the carried state to guide the next prediction without fixing it. ChainVLA reaches 62.8% average success on RMBench and 98.8% across four LIBERO suites, while removing Motion Tail or Progress Context reduces RMBench success to 11.2% and 3.0%, respectively. These asymmetric ablations are consistent with motion continuity helping preserve the observation stream from which task progress is inferred.

State estimation is a key component in model-based control of walking robots and, more broadly, applicable wherever hidden variables must be inferred. The Kalman filter is widely used to estimate floating-base position and velocity by fusing multiple sensing modalities. However, tuning noise parameters is challenging and typically requires expert knowledge. Moreover, fixed noise parameters are unsuitable for varying gaits and environments. We propose an online adaptation strategy for the process noise covariance matrix Q and the measurement noise covariance matrix R. Specifically, we introduce a filter residual and innovation-based covariance adaptation method for legged robot state estimation and evaluate it against a baseline approach relying on IMU and foot force measurements. The proposed adaptation is implemented within an Invariant Extended Kalman Filter (InEKF) fusing IMU and leg kinematics. Experiments on indoor and outdoor datasets with a Unitree Go2 quadruped show that adapting R is sufficient and improves accuracy by 25% for the trotting gait compared to the fixed-tuned InEKF. Finally, the proposed residual-based adaptation achieves comparable performance to the foot force approach, without requiring foot force measurements or additional parameter tuning.

Mobile manipulation is a key capability for embodied intelligence, enabling robots to accomplish complex multi-stage tasks in open-world environments. However, mobile manipulation poses two key challenges for vision-language-action (VLA) policies: At the data level, the efficient collection of high-quality whole-body demonstrations demands the coordinated control of both the mobile base and the robotic arms; at the model level, existing VLA models predominantly rely on local camera observations, whose limited field of view hinders global spatial understanding. To address these challenges, we develop a whole-body teleoperation system and a panoramic-aware VLA policy. The system enables coordinated control of a wheeled bimanual robot through a single VR interface and supports the acquisition of a real-world mobile manipulation dataset comprising 5.5 hours of multimodal demonstrations. Building upon this dataset, we propose PanoVLA, a panorama-aware vision-language-action policy for mobile bimanual manipulation. Built upon a Mixture-of-Transformers architecture, PanoVLA introduces global spatial context through dedicated panorama encoding and fusion modules, enabling effective integration of panoramic observations with language instructions and robot states for action generation. Evaluation on four real-world mobile manipulation tasks demonstrates that PanoVLA achieves an average stage completion rate of 91.3\% and an end-to-end success rate of 73.4\%, substantially outperforming local-view baselines. These results demonstrate that incorporating panoramic spatial context improves spatial understanding and closed-loop manipulation performance in mobile robots.

Visual representations of VLA models remain unreliable for spatially precise robotic manipulation. We uncover that vision encoders in VLAs also exhibit attention artifacts previously documented in generic Vision Transformers, and further show that, in embodied policies, these artifacts are closely associated with spatial perception capabilities acquired during post-training. As the encoder learns task-relevant information such as object location, depth ordering, and local geometry, limited global-token capacity causes part of this information to spill into low-information patch tokens. We introduce AtVLA, a framework that inserts learnable register tokens into the visual encoder. Trained end-to-end using only embodied data and the original action objective, these registers emerge as dedicated carriers of embodied spatial information, while the remaining patch tokens recover clean and spatially faithful attention distributions crucial for precise target localization and fine-grained contact. Clean attention restores reliable localization, but cannot recover geometric details lost in low-resolution observations. AtVLA therefore couples attention rectification with uncertainty-gated local refinement. The action expert samples multiple action chunks and estimates uncertainty from their disagreement; only for uncertain predictions, action-conditioned attention rollout identifies the task-relevant region, which is cropped, re-encoded at high resolution, and appended to the cached prefix for refined action generation. Across LIBERO, SimplerEnv, and a challenging single-view real-world benchmark, AtVLA improves the average LIBERO success rate from 94.2% to 98.4% and real-world success from 46.5% to 69.0%. The cropping is triggered on approximately 30% of replanning steps, resulting in only 1.4-1.6x the total computation of the base model under the representative deployment setting.

Reliable photovoltaic (PV) forecasts are needed for low-carbon energy systems, but newly deployed sites often have short, imperfect records. This makes standard day-ahead forecasting difficult: persistence and physical baselines can be sensitive to calibration and timestamp alignment, while single machine-learning models may capture only one structure in the data and overstate skill under non-temporal validation. We study this problem at a United Kingdom charging-station site, where PV forecast errors affect charging availability, storage scheduling, and downstream control. Using measured inverter output and publicly available meteorological inputs, we develop a deployment-oriented environmental-AI pipeline for day-ahead hourly PV forecasting. The pipeline corrects timestamp conventions, constructs leakage-safe solar-geometry and clearness-index features, adds short-term atmospheric context, and combines complementary predictors through validation-learned stacking. Against smart persistence, a clear-sky baseline that adjusts recent PV output using expected clear-sky irradiance, the best ensemble reduces daylight normalised RMSE by about 32% under random day-blocked evaluation and 9% under the stricter rolling-origin protocol. It also reduces daylight RMSE relative to the strongest individual machine-learning baseline by 6.6% and 6.4%, respectively. The results show that physics-aware stacking can support PV forecasts from limited site data, but its value depends on model class, evaluation protocol, and deployment context.

Whole-body tactile sensing is a prerequisite for humanoids that operate in contact-rich human environments, but conventional taxel arrays scale poorly with surface area, wiring complexity, and robot-specific curvature. We present a conformal electrical impedance tomography tactile skin fabricated through a geometry-adaptable additive-manufacturing workflow. A flexible conductive TPU layer forms a continuous sensing domain, while contact-induced coupling with conductive patches produces boundary voltage changes that are reconstructed using a one-step Gauss-Newton EIT solver. We first characterize the electromechanical design space of the layered structure and show that low-resistance contact-enhancement patches and a porous conductive TPU sensing layer improve sensitivity while preserving printability. We then validate contact localization on a planar prototype, a curved U-shaped prototype, and a qualitative iCub-face-shaped geometry. The curved sensor achieves a mean localization error of 6 mm over 18 contact positions without supervised post-processing. These results suggest that additively manufactured tomographic skins can reduce the morphology-specific redesign burden for humanoid tactile coverage and provide a practical route toward large-area contact sensing for human-centered deployment.

Developing deployable locomotion policies through conventional reinforcement learning often requires complex reward engineering and expensive training times. While differentiable simulation offers a highly efficient alternative, open-source tools capable of end-to-end transfer of these policies to physical hardware remain limited. This paper introduces Open-DiffLoco, an open-source framework for training deployable blind quadruped locomotion policies with differentiable simulation. The framework implements the Short-Horizon Actor-Critic (SHAC) algorithm in MuJoCo XLA (MJX) and trains a proprioceptive policy that transfers to real-world hardware. The deployed policy removes privileged actor observations, including base linear velocity, and does not rely on reference trajectories. It also uses a substantially simplified reward function, enabling the robot to discover walking patterns without the complex auxiliary rewards typically used in conventional reinforcement learning pipelines. When deployed on physical hardware (a Unitree Go2 quadruped), the trained policy tracks omnidirectional velocity commands with root-mean-square error below 0.2 m/s, reaches speeds above 1 m/s, and remains robust to uneven terrain and external physical disturbances, such as lateral pushes. Across the reported configurations, training uses under 6 GB of VRAM on a single NVIDIA GeForce RTX 5080 GPU and completes in approximately 20-60 minutes. As an algorithmic extension to SHAC, we propose Jacobian-Augmented Value Estimation (JAVE), which supervises the critic Jacobians to improve early first-order policy-gradient training. To our knowledge, Open-DiffLoco is the first open-source framework for training deployable locomotion policies using differentiable simulation. Deployment videos and source code are available at: https://diffloco.martin-opat.com/

Cross-embodiment dexterous grasping aims to synthesize stable grasps across heterogeneous multi-fingered hands with little or no embodiment-specific tuning. Existing interaction-centric methods achieve promising results, but their object representations often underrepresent local surface geometry, while their robot descriptors do not explicitly encode both robot morphology and kinematics. We propose MANGO-Grasp, an anisotropic interaction framework that represents objects as geometry-oriented 3D Gaussian primitives and robot hands as surface keypoints encoded into morpho-kinematic descriptors. The object primitives are adaptively allocated by geometric complexity and shaped as surface-aligned plates with outward normals, encoding local geometry. Mahalanobis fields over keypoint--primitive pairs serve as interaction prediction targets during training and as optimization guidance for grasp realization at inference. These fields rise sharply for displacement along the surface normal but only gently within the tangent plane, matching the directional structure of contact. Grasps are realized with one shared optimization formulation and hyperparameter setting across all embodiments. On the CMAP and MultiGripperGrasp benchmarks, MANGO-Grasp outperforms the strongest seen-hand baseline by up to 8.24 percentage points in simulation. It also transfers zero-shot to the unseen SharpaWave hand, improving over the strongest zero-shot baseline by up to 16.57 percentage points, and achieves 86% success in real-world experiments. The code and additional materials will be made available upon publication at https://connor-zh.github.io/MANGO-Grasp/.

Existing indoor layout generators produce globally plausible layouts yet may retain local violations such as collisions, out-of-bounds placements, obstructed openings, and blocked circulation. Most prior work focuses on full-scene synthesis or scene-level optimization, with limited support for identifying responsible objects and locally repairing affected regions. We present Roomer, a reflective repair framework that casts these violations as sparse, object-grounded repair problems. Roomer encodes layouts as ``RoState'' and uses ``RoReview'' to bind measured violations to implicated objects. A geometry-conditioned vision-language model planner proposes a structured local edit, while a deterministic solver validates it and generates a finite set of candidate edits when needed. Each candidate is committed only if full-scene verification confirms that it resolves the target violation without new hard violations or broken protected constraints. We train the planner on Roomer-CC, a controlled-corruption dataset that pairs faulty layouts with object-grounded violation evidence and known-feasible inverse StatePatches. Since existing benchmarks rarely assess whether physically valid layouts are usable, we introduce Roomer-Eval to assess distributional quality, physical validity, and practical usability. Experiments show that Roomer repairs residual violations while preserving valid regions, improves physical validity and usability, and transfers across external generators.

Passive overlay communication for batteryless devices is an important enabling capability for next-generation vehicle-to-everything (V2X) networks. However, enabling reliable passive payload delivery without occupying additional spectrum remains challenging, since overlay signaling must be embedded into short and time-varying vehicular packets while preserving the decodability of the legacy host transmission. This paper investigates a packetized batteryless V2X overlay architecture in which a dedicated short-range communications (DSRC)-based packet simultaneously carries conventional V2X data and a passive overlay payload. A compact PHY-layer model is developed to characterize the coupled effects of attenuation depth, embedded-bit rate, and legacy modulation and coding scheme (MCS) on host-link and passive-link reliability, as well as packet-level embedding feasibility. We then formulate a sum-throughput maximization problem that jointly accounts for the legacy packet error rate and passive decoding error rate. We further propose a multi-agent reinforcement learning (MARL)-based adaptive parameter-selection method. Simulation results show that the proposed MARL controller achieves stable convergence and improves the average throughput by 15\%, demonstrating the effectiveness of throughput-driven PHY adaptation for batteryless V2X overlay communications.

World Action Models generate fixed-horizon action chunks through iterative denoising, creating substantial inference latency that can cause pauses, stale actions, and discontinuities during robotic execution. We present an empirical study of asynchronous deployment strategies that overlap model inference with action execution to enable responsive and smooth control. We compare six strategies, including synchronous execution, pure asynchronous switching, post-hoc action blending, denoising-time blending, inference-time velocity guidance, and prefix-conditioned generation, on a 10 Hz bimanual robot. Evaluation combines offline trajectory analysis with online experiments across dynamic manipulation, precision-critical placement, and long-horizon tasks. Our results identify accurate temporal alignment between observations, predictions, and executed commands as a fundamental requirement. Alignment errors produce persistent chunk-boundary discontinuities that cannot be corrected through blending alone. With proper alignment, direct action weighting provides a simple and smooth baseline but sacrifices accuracy in precision-critical tasks. Inference-time velocity guidance fails to reliably constrain committed actions on our platform. In contrast, prefix-conditioned generation achieves the best overall balance between task performance, execution speed, and trajectory smoothness by learning consistent action continuations during training. These findings clarify the practical trade-offs among asynchronous deployment strategies and provide guidance for deploying high-latency World Action Models in real-time robotic systems.

Robot learning is splitting into two bets: policies that bake competence into frozen weights (vision-language-action, or VLA, models), and agents that write and refine their own executable skills as code. This survey organises the field around that axis of weights versus skills. Its central analytical contribution is a deep-dive that arranges code-as-policy methods by their degree of self-improvement, from zero-shot program synthesis, through closed-loop self-repair and persistent skill memory, to the sparsely populated cell in which execution feedback, skill memory, and evolutionary search combine into one open-ended loop; only a few very recent systems (for example ASPIRE, ENPIRE, and RoboClaw) occupy that cell. We map the complementary "skills" pole, from unsupervised reinforcement-learning skill discovery to large-language-model skill libraries, and show that the word "skill" is used in at least five distinct senses, of which only the code sense self-improves without gradient updates. We then connect the taxonomy to the emerging skill economy: commercial robot-skill marketplaces now distribute one-tap skills across robots but ship only static playback, which surfaces open problems of adaptation, cross-embodiment portability, provenance, safety verification, composition, and standardisation. This is a deliberately focused survey. Rather than cataloguing the field exhaustively, it examines 77 representative systems across six technique families through one taxonomy and a set of contrast tables, and it supplies operational definitions of the self-improvement mechanisms together with a statement of what each family cannot do.

Humanoid teleoperation for demonstration collection requires coordinated whole-body motion, continuous dexterous hand control, and viewpoint control. Existing systems either simplify hand commands or depend on dedicated wearable sensors for fine-grained hand motion. We introduce Teleopit, a full-embodiment teleoperation system that maps body, hand, and head signals from VR to a humanoid body, configurable dexterous hands, and a 2-DoF active vision module. A history encoder and failure-aware rewind sampling improve the motion tracker on both motion-capture and live VR references. An optimization-based hand retargeter combines normalized finger directions, fingertip closure, and thumb-frame alignment to map human hand motion to different dexterous hands without tuning hand-specific objective or solver hyperparameters. Component experiments evaluate tracking success rate and retargeting behavior, while real-robot teleoperation demonstrates coordinated locomotion, manipulation, and viewpoint control. ACT and GR00T N1.7 policies trained on 96 successful demonstrations collected with Teleopit achieve task success rates of 90.0% and 95.0%, respectively, when deployed on the humanoid. The project page is available at https://botrunner64.github.io/teleopit-page.

Vision-Language-Action (VLA) models have shown strong generalization in robotic manipulation, yet complex contact-rich tasks often benefit from multi-camera observations that jointly capture the end effector, objects, and targets under occlusion. Existing multi-camera VLAs usually concatenate view tokens, leaving action representations weak in metric depth and inconsistent across cameras. We introduce Multi-View Unified Camera Fields (MVUCF), a training-only framework that forms a shared action-facing latent field across views. A coordinate-query depth objective makes metric depth recoverable, while a preprocessing-aware correspondence objective aligns tokens observing the same physical point from different cameras. Both directly shape the hidden states consumed by the action module. After geometry injection, depth, camera calibration, and auxiliary heads are removed, so deployment uses the original RGB-only graph with no extra inference FLOPs. Held-out probes confirm stronger depth recovery and cross-view matching. Under matched GR00T-N1.6 settings, MVUCF reaches 98.9% on LIBERO, improves LIBERO-Plus by 22.4 points, and raises success by 23.3 points across six RoboTwin tasks spanning three action families: touch, move-and-place, and contact interaction. Real-world humanoid experiments further provide evidence of its practical effectiveness under RGB-only deployment.

Fusing tactile signals has proven effective for contact-rich manipulation, enabling robots to perceive contact states and adapt to rapidly changing physical interactions. Yet effectively integrating tactile feedback into dexterous manipulation remains underexplored. In this work, we introduce ReTouch, a vision-language-action model (VLA) that supports contact-rich dexterous manipulation through tactile predictions continually refined online using execution-time feedback. ReTouch builds on two main innovations for tactile representation and closed-loop action generation. First, its Tactile-Patch Encoder represents tactile observations as structured tactile patch features that preserve finger identity and local contact structure, providing contact cues for fine-grained dexterous control. Second, its high-frequency action module jointly predicts future tactile states and action chunks and refines both using incoming tactile feedback during execution. This closed-loop refinement keeps tactile predictions aligned with evolving physical interactions, enabling responsive action correction and improving robustness to contact changes and execution errors. We further introduce XHT-Dataset, comprising 900 real-world demonstrations across seven contact-rich tasks collected on an XHand--UR7e platform, and evaluate ReTouch through closed-loop real-robot experiments. ReTouch surpasses the strongest baseline by 18.4 and 23.8 percentage points in average success rate under standard and challenging conditions, respectively, demonstrating its effectiveness and robustness.

Target-oriented vision-and-language navigation (VLN) on aerial platforms is attracting growing attention for missions such as disaster rescue, infrastructure inspection, and security patrol. In this task, an unmanned aerial vehicle (UAV) needs to locate targets given only a concise description of their appearance and surroundings. This requires global exploration and grounding as well as collision-free close-range approach, two interleaved processes difficult to reconcile within a single agent. Most existing methods transfer the ground VLN paradigm to a low-altitude UAV and compensate for its inefficient exploration with external assistance. A recent attempt deploys two UAVs at complementary altitudes yet still relies on privileged information and trains its two agents independently, precluding any mutual adaptation essential for cooperation. Here we propose CoNav-UAV, which explicitly models the task as a Stackelberg game between a high-altitude leader and a low-altitude follower, with the system operating on onboard visual and linguistic inputs alone. To solve this game, we introduce Iterative Stackelberg Learning. The leader's high-level vision-language reasoning is refined via memory-based in-context learning, while the follower's precise motion control is updated via DAgger-style expert distillation. The alternation drives both agents toward a Stackelberg equilibrium. CoNav-UAV consistently outperforms single- and dual-agent baselines across three high-fidelity urban scenes from the AerialVLN benchmark. Success rate improves by up to 30.8 points on the learning scene, and 9.0 points under cross-scene transfer while using about 3x less adaptation data. Further analyses validate the complementary gains of the leader and follower updates and reveal robust gains yet distinct learning dynamics across VLM backbones.

Multi-link aerial robots can actively deform their articulated structures during flight, giving them strong potential for aerial manipulation. However, they still face substantial challenges in contact-rich aerial manipulation tasks such as surface sliding, which requires both disturbance robustness and compliance to uncertain surface geometry. Force-control strategies such as impedance and admittance control are commonly employed to address these requirements. Although impedance control can provide disturbance-resistant interaction and admittance control can offer compliant adaptation, their opposite force--motion causalities prevent their simultaneous implementation when applied through the same actuation source, such as the rotor thrusts used by conventional aerial robots. To overcome this limitation, we propose a hybrid impedance--admittance control strategy for a multi-link aerial robot. The articulated morphology enables a functional separation of force and motion regulation across joint and rotor actuation sources. In this framework, admittance behavior is generated through joint angle regulation to enhance adaptive interaction, while impedance behavior is achieved by modulating rotor thrust to regulate the sliding motion. This structural coordination allows the robot to leverage the complementary strengths of both control paradigms. As a result, the multi-link aerial robot achieves resilient and adaptive surface sliding. Experimental results demonstrate robust and compliant sliding performance on unknown surfaces.

Maintaining a stable pose during physical interaction is a significant challenge for aerial robots, often limiting their use in contact-based tasks. This paper presents a novel uncrewed aerial vehicle (UAV) platform designed to transition from unconstrained flight to a stable, constrained work platform via environmental anchoring. Our system comprises: 1) a multirotor with a tilt-rotor mechanism that decouples pitch from forward motion, enabling stable hover at non-zero pitch angles, and 2) a novel underactuated, cable-driven, prismatic gripper featuring compliance to adapt to irregular geometries, designed to stabilize the UAV by anchoring it to its environment. We present the design and prototyping of the complete system and validate its performance through a series of real-robot flight tests. Results demonstrate that anchoring significantly improves stability for interaction tasks, reducing positional drift RMSE by over 95% compared to a free-flight baseline, even under windy conditions. The anchored system can withstand longitudinal reaction forces up to 75N while maintaining a stable pose. Furthermore, across a range of target geometries and orientations, the system demonstrated consistent stability with a positional drift RMSE that never exceeded 3mm. These results establish the viability of our approach for complex physical interaction tasks, such as sampling tree health by drilling or sensor installation in hard-to-reach locations. Watch our UAV at: https://youtu.be/HDQ8S4ZW3Ls

Recent advances in robot learning for manipulation have increased the importance of collecting real-world demonstration data. However, existing robotic systems primarily focus on end-effector manipulation, making it difficult to teach and execute manipulation tasks involving body-surface contact with the arms and chest. This paper presents TWINS (Tactile Wearable Isomorphic Arm Networked System), a robotic system for manipulation involving body-surface contact. TWINS consists of a Wearable Dual-Arm Device, which is worn by the operator, and an Isomorphic Robot with the same joint configuration and external dimensions. Distributed tactile sensors embedded in the chest and arms enable the measurement of body-surface contact synchronized with joint motion. Using the Wearable Dual-Arm Device, we collected demonstrations for four manipulation tasks involving body-surface contact. We then trained imitation learning policies using the collected demonstrations and deployed them on the Isomorphic Robot, enabling manipulation guided by body-surface tactile observations. Experimental results demonstrate that TWINS provides a unified robotic system for demonstration, learning, and execution of manipulation involving body-surface contact. https://mmurooka.github.io/twins-project-page/

This paper addresses the sim-to-real gap in dynamics arising when a parallel-link mechanism is represented by a serial-tree surrogate in simulation. Conventional Jacobian-based state and torque mappings preserve consistency with the kinematic and virtual-work relations but do not account for the coordinate-induced redistribution of actuator inertia and damping and the linkage inertia omitted during serial-tree reduction. To address this gap, Simulator-Side System Normalization (S3N) is proposed to normalize the serial-tree simulator's effective dynamics while preserving its tree topology. S3N-Act incorporates actuator inertia and damping into the serial-coordinate dynamics through coordinate transformation, whereas S3N-Full restores residual linkage inertia by separately identifying actuator- and leg-level frequency responses. In the 2-DoF validation, S3N-Full reduced the joint-position and torque RMSEs by 80.9% and 82.1%, respectively, relative to the Jacobian-mapping baseline. During pitch-in-place motion, S3N-Act and S3N-Full reduced the RMSE of the ground reaction force norm by 65.1% and 62.4%, respectively. During circular locomotion, S3N-Full reduced the phase-averaged, command-normalized sim-to-real gap from 17.3% to 9.9%. These results show that simulator-side normalization improves motion- and force-level sim-to-real consistency. It enables policy training in a serial-tree framework with hardware-consistent dynamics that better represent the physical parallel-link mechanism.

Biological wet-lab protocols are written for trained researchers and often leave routine operations, state-dependent conditions, and contextual parameters implicit, making them difficult to translate into robot-executable actions. We present ProtoAct, a structured protocol-grounding framework that converts free-form biological procedures into state-aware, embodiment-ready action sequences. ProtoAct uses ProtoRAG to retrieve manually annotated examples for context-sensitive parsing, employs RefineChecker to detect and revise missing or inconsistent steps, and applies ActSchema to map the refined procedure into constrained JSON function sequences. We further introduce BioP2E, for which we manually annotate 22 cell-culture protocols into 258 monitoring conditions, 910 executable subtasks, and 962 grounded action calls. Evaluation across seven large language models demonstrates that ProtoAct can be effectively instantiated with different backbones. Ablations confirm that retrieval, posterior checking, and schema constraints make complementary contributions. The parsed subtasks further support demonstration collection and VLA model training, enabling successful execution in both simulation and real-robot settings. ProtoAct thus provides a practical interface between biological protocol understanding and embodied robotic execution.

LLM-based multi-agent coordination faces a fundamental trade-off between efficiency and adaptivity in dynamic environments. Existing approaches typically rely on repeated LLM invocations or multi-round communication to adapt decisions during execution, introducing substantial latency and making coordination vulnerable to asynchronous progress and environmental changes. Conversely, one-shot planning reduces coordination overhead but produces open-loop plans that can quickly become stale or fail when actions depend on other agents and the environment. We introduce SyncPlan, a plan-execute-correct framework for long-horizon coordination through explicit synchronization and adaptive correction. Given the state and team-level task, a centralized LLM coordinator generates per-agent action chains in a single planning call. During execution, explicit wait primitives and deadlock detection enforce inter-agent and agent-environment dependencies, while a lightweight Plan Staleness Detector continuously assesses the remaining plan and triggers replanning when environmental changes invalidate its assumptions. We further optimize the coordinator through SFT and planning-oriented RL with dense task progress and outcome-level execution feedback. Experiments on the public Overcooked benchmark and the complex Honor of Kings environment show that SyncPlan achieves state-of-the-art task success rates while using less than 0.05% of the wall-clock runtime compared with existing LLM-based coordinators. Code and datasets will be made publicly available.

This paper studies feedback Nash equilibrium (FBNE) seeking for multi-agent trajectory planning in nonlinear dynamical systems with unknown agents' objectives and state-dependent inter-agent coupling. While dynamic game theory provides a principled framework for such problems, existing approaches typically assume fully rational agents with known objectives or rely on fixed regularization, limiting their ability to capture bounded rationality and spatially varying interaction intensity in safety-critical settings. To this end, we propose a KL-regularized dynamic game with a state-dependent weight that adaptively balances optimality and behavioral priors. To infer unknown cost parameters from demonstrated behaviors, we develop a context-aware inverse game module based on maximum-entropy inverse reinforcement learning with physics-informed regularization, ensuring structural consistency with the forward game. We establish per-iteration well-posedness of the regularized local game and show that the adaptive weighting function remains Lipschitz continuous under bounded nominal-trajectory updates. Numerical simulations and multi-robot experiments on cooperative navigation and merging scenarios validate the effectiveness of the proposed framework.

Affordance-guided imitation learning has shown impressive performance in robotic manipulation tasks by compressing visual perception into task-specific geometric constraints (e.g., fixed contact points). However, the commonly used static affordances can become inconsistent in precision-critical tasks or under object location perturbations, leading to post-contact trajectory drift. To address this issue, we propose AffordTrajDP, a dynamic framework that constructs affordance trajectories via object-centric temporal propagation to guide the progressive manipulation process. Specifically, given an RGB-D observation, our core insight is that a retrieved anchor affordance, which captures the desired contact point between the end-effector and the target object, can be propagated forward via affordance propagation, using the object's SE(3) pose as a natural propagation medium, to yield an affordance trajectory that provides temporally consistent, state-aware guidance throughout execution. AffordTrajDP achieves 70.0% average success rate on ManiSkill3, outperforming strong baselines by up to 17.8%. Real-world experiments on Galaxea A1 and UR7e robotic arms, covering StackCube, PickCup, AdapterInsertion, Ring-on-Peg, Put-in-Bowl, and USB Insertion, further validate robustness under object placement variations and appearance changes, with seen and unseen object instances evaluated on Galaxea A1, and ablations confirm the contribution of each proposed component.

Humanoid robots, with their human-like shape and multi-tasking capabilities, are well-aligned with human-dominated workplaces, like those in civil and construction engineering, where they could collaborate with human workers or autonomously perform physically demanding and hazardous tasks. Despite this promise, limited research has explored how to endow these robots with the practical capabilities needed to perform construction tasks. To this end, this study proposes a novel perception-and-action system that enables humanoid robots to learn and perform construction tasks from worker demonstrations. This system contains two deep networks: Humanoid-PoseNet, which extracts human postures and translates them into mechanically feasible poses for a humanoid robot; and Humanoid-ActionNet, which learns robot-executable actions based on these translated poses. Experimental results demonstrate that the humanoid robot reliably executed eight construction-related actions, achieving an average motion-tracking error of 82.45 mm MPJPE (Mean Per Joint Position Error). This work provides an early step toward deploying humanoid collaborators in construction.

Recent Vision-Language-Action (VLA) models achieve promising performance in robotic manipulation, typically measured by success rates aggregated over predefined object configurations, an evaluation that implicitly assumes spatially uniform competence across the workspace. However, this assumption does not hold: even with the instruction and every other scene factor held fixed, merely relocating a task-irrelevant distractor can sharply raise the failure probability within localized, spatially coherent regions, which we term Positional Blind Spots (PBS). In this paper, we propose a two-stage black-box framework to uncover and mitigate PBS. During the uncovering stage, we grid the workspace and apply a one-sided log-likelihood-ratio test to localize PBS cells with significantly elevated risk. During the mitigation stage, we fine-tune the policy via LoRA on demonstrations collected from these PBS regions, improving competence there while largely preserving performance across the rest of the workspace. We evaluate our framework on five state-of-the-art VLA policies across two benchmarks, and find that PBS are pervasive and spatially concentrated in all of them, with failure rates up to 0.58. Our search strategy achieves an average F1-score of 0.678, outperforming random search and adaptive sampling baselines by 0.268 and 0.178, respectively. Guided by the discovered regions, targeted fine-tuning reduces the overall failure rate by 40.00%--85.19%.

Vision-language models (VLMs) are emerging as a key component of embodied intelligence, with growing applications in auto-labeling and end-to-end autonomous driving. However, existing approaches for improving spatiotemporal reasoning in VLMs often rely on complex preprocessing pipelines, expensive human annotations, or synthetic data, which limit scalability and introduce potential sim-to-real gaps. Moreover, although these methods have improved spatiotemporal understanding, they still lack strong metric reasoning capabilities for dynamic scenes, such as estimating object motion in real-world units. Prior work has explored LiDAR-based metric depth supervision to enhance spatial perception, but it does not directly address temporal reasoning. We introduce STAR-VLM, an automotive radar-supervised framework that enhances spatiotemporal VLMs with motion reasoning and metric velocity estimation for autonomous driving. Automotive radar is a low-cost and widely deployed sensor that provides complementary spatiotemporal supervision through range and Doppler measurements. By leveraging these measurements as label-free ground truth during training, STAR-VLM improves the metric spatiotemporal reasoning ability of VLMs. Through experiments on driving scenarios, we show that STAR-VLM achieves state-of-the-art performance on both motion classification and metric velocity estimation, outperforming even task-specific methods designed for each task. These results highlight automotive radar as a scalable and cost-effective source of supervision for building metric-aware spatiotemporal VLMs for real-world autonomous driving.

Distributed target tracking in multi-agent networks plays a critical role in cooperative sensing and autonomous navigation. However, it faces significant challenges in highly dynamic and adversarial setups. This study aims to enhance the resilience of decentralized target tracking algorithms against measurement faults and cyber-physical threats, especially false data injection attacks. We propose a consensus-based estimation algorithm that integrates a nearly-constant-velocity model with saturation-based filtering to suppress impulsive measurement variations and promote robust, distributed state estimation. To counteract adversarial conditions, we incorporate a dynamic false data injection detection and isolation mechanism that uses innovation thresholds to identify and disregard suspicious measurements before they can degrade the global estimate. The effectiveness of the proposed algorithms is demonstrated through a series of simulation-based case studies under both benign and adversarial conditions. The results show that increased network connectivity and higher consensus iteration rates improve estimation accuracy and convergence speed, while properly tuned saturation filters achieve a practical balance between fault suppression and accurate estimation. Furthermore, under localized, coordinated, and transient false data injection attacks, the detection mechanism successfully identifies compromised agents and prevents their data from corrupting the distributed global estimate. Overall, this study illustrates that the proposed algorithm provides a simplified fault-tolerant solution that significantly enhances the accuracy and resilience of distributed target tracking without imposing excessive communication or computational burdens.

Substantial research efforts have been devoted to the design of data-driven controllers; however, comparatively less is known about their statistical performance and fundamental limitations. This contribution develops a statistical decision framework for data-driven control, in which a controller is evaluated by its risk, defined as the expected performance degradation relative to the oracle model-based controller, and by its average risk over the parameter space. Within this framework, we propose a collection of design principles for data-driven controllers. We further derive lower bounds on risks by combining the bias-variance decomposition with the Cramér-Rao inequality. In particular, the optimal bias that attains the lower bound for the average risk is determined by calculus of variations, thereby making the bias-variance tradeoff in data-driven control explicit. Moreover, the derived bound reveals a ``waterbed'' effect in data-driven control: any improvement in risk relative to the lower bound over one region of the parameter space must be compensated by deterioration elsewhere. We illustrate the proposed framework on two canonical data-driven control problems: optimal feedforward control and the linear quadratic regulator benchmark. By comparing several representative data-driven controllers with the derived lower bounds, we sharpen the statistical interpretation of existing methods and reveal quantitative limitations that no controller design can avoid.

Mixed complementarity problems (MCPs) arise as the first-order optimality conditions of nonlinear programs and noncooperative games, and provide a natural formulation for multi-agent trajectory optimization problems that appear throughout robotics. The dominant solver for problems of this form is PATH, which offers strong performance on robotics problems but remains closed-source. We present MixedComplementarityProblems.jl, an open-source, pure Julia implementation of an interior point method for parametric MCPs that: (i) matches PATH's reliability on standard benchmarks, (ii) natively supports batched, parallel processing of many parameter instances, either across CPU threads or on an NVIDIA GPU, and (iii) supports efficient automatic differentiation of solutions with respect to problem parameters. On a multi-agent lane-change trajectory game representative of robotics planning problems, our CPU-multithreaded batched solver clears a batch of parametric instances ~100x faster than sequential calls to PATH. A GPU backend, running the same solver implementation unmodified, also clears these batches far faster than PATH, but does not outperform the multithreaded CPU on this problem; the GPU pulls ahead only once each per-instance KKT system grows large, and we characterize this regime dependence. We describe the solver's interior point formulation, the abstraction that lets a single solver implementation run unmodified across dense, batched-sparse, and single-large linear-algebra backends, and report benchmarks against PATH on both randomly generated quadratic programs and trajectory games.

The rapid integration of inverter-based resources (IBRs) is reducing system inertia across U.S. power grids, raising concerns about frequency stability following large contingencies. This paper presents a simulation-based assessment of critical inertia, defined as the minimum system inertia required to prevent first-stage under-frequency load shedding (UFLS) after the largest credible contingency, across the three major U.S. interconnections: Eastern Interconnection (EI), WECC, and ERCOT. Reduced-inertia scenarios are created by progressively replacing synchronous generators with IBRs, and dynamic simulations are performed using full-scale PSS/E and PowerWorld models. The results show that ERCOT reaches critical inertia at approximately 58 percent IBR penetration, compared with above 90 percent for WECC and approximately 67 to 68 percent for EI. Current IBR shares in the U.S. portions of EI, WECC, and ERCOT are 16 percent, 33 percent, and 44 percent, respectively, indicating varying proximity to critical inertia thresholds. These findings highlight the importance of full dynamic simulations to accurately estimate critical inertia and guide transmission planning under high renewable penetration scenarios.

Standard multifingered grasp execution architectures plan a collision-free trajectory to a selected grasp pose and track it with a feedback law. Execution-time object pose uncertainty or perturbations may invalidate the planned trajectory, forcing a costly replanning step. We present Grasp Distance Fields (GDFs), smooth softmin distance fields over grasp configurations in the arm-hand configuration space. Our controller executes a grasp by following the negative GDF gradient with a stationary feedback law, eliminating the need for a planner, stored trajectory, or grasp selection. For safety, we filter the resulting command through a CBF-CLF quadratic program (QP), constraining self-collision, workspace, object, and obstacle clearance, reporting impeded progress as explicit slack. We prove that the softmin tracks the true set distance within $\log N/ρ$ for $N$ candidates and smoothing parameter $ρ$, and that the filtered closed loop renders the safe set forward invariant. Since no smooth field captures hand-object contact switching, we switch between hand closure and grasp retention modes using hysteresis at a pregrasp configuration, trading object collision avoidance for contact admission. A wrench-quality CBF then bounds the realized grasp's force-closure margin within a prescribed tolerance of its value at hold onset. We evaluate in kinematic simulation on a fixed-base arm and a Unitree G1 humanoid, both fitted with the same underactuated hand. Our controller navigates cluttered and dynamic scenes to safely reach for, grasp, and lift 46 of 50 test objects spanning primitive, household, and adversarial classes. Across the 46 lifts, the executed grasps retain a median 94% of their synthesized quality margin, at 0.09 ms QP solve time per 20 ms control step. Project page: www.clintonenwerem.com/gdf.

Deploying billion-parameter Vision-Language-Action (VLA) policies on mobile robots creates a systems conflict: semantic reasoning benefits from cloud GPUs, whereas closed-loop control must respond locally despite network delay and jitter. Existing hierarchical and asynchronous policies improve throughput, but their slow-path representations can still arrive stale or require explicit scheduling and delay cues. We introduce CloudEdgeVLA, a cloud-edge policy that treats temporal misalignment as a representation-learning problem. A cloud VLA encodes delayed observations into slowly varying task features, while a lightweight edge head combines the latest available cloud feature with current local vision. During training, current and randomly delayed frames are paired with the same current action target in fresh and stale paths. This objective encourages the cloud representation to preserve task-level information while the edge path supplies state-sensitive corrections. Across four LIBERO suites, CloudEdgeVLA retains 63.8--78.0% success with a 40-step uniform-delay window, whereas VLASH reaches at most 6.4% and the evaluated single-path baselines at most 3.0%. By removing blocking synchronization from the control loop, the design offers a practical route to scalable VLA deployment in which cloud models can grow while edge computation remains lightweight and responsive.

Robot policies are becoming increasingly general, with vision-language-action (VLA) models enabling a single policy to execute diverse tasks specified in natural language. Safe deployment, however, requires adapting not only to new tasks but also to varying safety requirements across users, environments, and applications. Existing safety filters remain largely constraint-specific and thus must be redesigned or relearned when safety requirements change. In this paper, we investigate language-conditioned safety filtering, in which a Hamilton-Jacobi safety actor and critic are conditioned on language-specified constraints. We evaluate this formulation across pick-and-place, table-wiping, and block-stacking tasks in the vision-based setting, examining its ability to enforce language-specified constraints and transfer to unseen constraint instances within the evaluated constraint families. Our experiments provide evidence that language-conditioned safety filters reduce constraint violations and exhibit partial transfer to unseen constraint instances.