Research

• Compositionality in RAG pipelines1 paper raises it · 1 addresses it

  Combining individually strong RAG components often fails to improve overall performance, and gains may not transfer across retrieval quality and answer faithfulness.

  • Retrieval Augmented Generation Compositionalityraises · addresses
• Request execution in streaming video generation1 paper raises it · 0 address it

  Models struggle to reliably follow incremental multimodal requests while maintaining state consistency and visual fidelity.

  • StreamEval: Evaluating Interactive Streaming Video Generation Modelsraises
• Brain-body mismatch in recursive self-improving agents1 paper raises it · 1 addresses it

  Evolving execution harnesses may demand behaviors the model has not learned or restrict existing capabilities.

  • CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow?raises · addresses
• Length generalization in state tracking1 paper raises it · 1 addresses it

  Models trained on short sequences often fail to generalize to much longer sequences without explicit progress alignment.

  • PAR: Progress-Aligned Recurrence for Length Generalization in State Trackingraises · addresses
• Evidence-grounded evaluation for audio-visual reasoning1 paper raises it · 1 addresses it

  Existing benchmarks often allow models to answer correctly without analyzing necessary audio-visual evidence, weakening reliability.

  • EvaOmni: An Evidence-Grounded Benchmark for Audio-Visual Reasoningraises · addresses

Named by an AI from these papers' abstracts.

• arXiv cs.ROOct 14

  What Matters for World Modeling in Vision-Language-Action Models?

  Qisen Ma, Yixiang Chen, Yuan Xu, Peiyan Li, Jiabing Yang +8

  Integrating world models into Vision-Language-Action models (VLAs) has emerged as a promising direction, enabling policies to anticipate future dynamics beyond reactive observation-to-action mapping. However, existing approaches vary simultaneously in their world representations, integration paradigms, and policy backbones, making it difficult to determine what truly matters for effective world model integration. To disentangle these factors, we categorize existing efforts into four representative paradigms, namely Implicit Future Alignment, Explicit Future Query, MoT Future Expert, and External World Predictor. We instantiate each within a unified framework and systematically evaluate these paradigms with diverse world representations and policy backbones across in-domain, out-of-domain, long-horizon, dynamic, few-shot, and real-world settings. Our study yields four main findings: 1) world representation effectiveness is paradigm-dependent, whereas 2D semantic representations offer robust, cross-paradigm gains; 2) no single paradigm is universally optimal, yet the decoupled MoT Future Expert offers competitive overall performance, particularly in OOD robustness; 3) world…

  PDF

  Summarize this
• arXiv cs.CVOct 14

  PhyMetric: Diagnosing Physical Plausibility in Text-to-Video Generation via Scene-Level QA

  Siwei Meng, Xuerou Cai, Rui Hu, Ping Liu

  Text-to-video (T2V) generation models can produce visually realistic videos that nevertheless violate the physical consequences implied by the prompt and scene. Existing benchmarks largely rely on holistic plausibility scores or broad quality dimensions, leaving it unclear which physical behavior is being tested and why a model fails. We introduce PhyMetric, a scene-level diagnostic benchmark that recasts physical plausibility evaluation as verifiable, scene-grounded question answering. PhyMetric contains 1{,}115 scenes spanning 50 physical categories across six domains, paired with 8{,}920 answer-balanced binary questions over 12 evaluation dimensions. Each scene is associated with eight targeted yes/no questions and a structured physical prior specifying the governing law, expected temporal sequence, observable signatures, and common failure modes. At evaluation time, a vision-language model (VLM) judge answers each question from question-aware sampled frames conditioned on the prior, and correctness is hierarchically aggregated into scene, dimension, physical category, and model scores. Experiments on four open-source T2V models and three VLM judges show that scene-specific…

  PDF

  Summarize this
• arXiv cs.DCOct 14

  JointServe: Serving Multi-Model Agents on Memory-Constrained GPUs

  Daijin Hu

  LLM agents are evolving from a single model calling tools to planners that orchestrate multiple expert models whose capabilities live in distinct weights. This shift breaks a hidden assumption of today's serving systems: that the working set of models fits in GPU memory. On the memory-constrained GPUs typical of private and edge deployments, naive whole-model swapping serializes weight transfer with computation and, without host-backed KV restoration, discards session state at a switch. We present JointServe, a serving system that treats weight shards and KV blocks as residency objects competing for one unified memory budget. JointServe exploits three structural properties of agent workloads (directed calls, gapped execution, and dual-state residency) to prepare models before they are called: an intent signal parsed from the caller's own output stream triggers prefetch during generation, partial k* residency reduces switching to incremental layers, and a preemptive transfer scheduler protects live-session KV with a guaranteed bandwidth share. A carve-out control step adjusts the weight/KV allocation as workload demands change. Evaluations show that JointServe reduces session…

  PDF

  Summarize this
• arXiv cs.AIOct 14▲ 57

  ActionOPD: Distilling Decisions, Not Tokens

  Huining Yuan, Xiangmin Yi, Zelai Xu, Zixu Wang, Xinlei Chen +3

  Distilling capable language-model agents into smaller models requires aligning the student's behavior with the teacher's. This raises a fundamental question: at which states and at what granularity should this alignment be performed? Offline imitation provides dense but token-specific supervision and can suffer from state-distribution mismatch. Token-level on-policy distillation (OPD) addresses this mismatch but typically requires teacher token probabilities and compatible tokenization. We argue that semantic actions provide a natural unit of supervision for agent distillation. We introduce ActionOPD, an action-level on-policy distillation method motivated by maximizing expected teacher-action likelihood at student-encountered states. At each such state, ActionOPD samples actions from both the student and a black-box teacher, rewarding student actions according to their semantic agreement with the sampled teacher actions. Through policy optimization, ActionOPD aligns the student's action-level behavior with the teacher's decisions without requiring reproduction of teacher-written reasoning or access to teacher token probabilities. Training Qwen3-4B and Qwen3-8B with two teachers…

  Code
• arXiv cs.AIOct 14

  One Monitor Does Not Fit All: Formalism-Agnostic Runtime Verification for Agentic AI

  Angelo Ferrando

  Agentic AI systems combine model decisions, mutable state, tool calls, and delegation to external agents. Monitoring them requires observations from different execution boundaries, while prevention requires control before an operation produces its effect. We present AgenticRV, a runtime-verification architecture that separates these boundaries from the monitoring formalism. Adapters emit canonical events; each monitor receives its own projection and abstraction; and a decision policy uses normalised verdicts to select a reaction supported by the responsible adapter. We formalise when observations suffice, monitor replacement preserves decisions, and blocking prevents unsafe effects. We also distinguish rejected attempts from completed effects, avoiding the assumption that a blocked request must leave every monitor in a permanent violation state. The Python implementation supports LangGraph, MCP, and A2A. A controlled LangGraph experiment combines an external Reelay temporal monitor with a programmatic report monitor; ablations show why both are needed for the selected requirements. Differential checks and cross-adapter tests provide further integration evidence. An LLM-driven…

  PDF

  Summarize this
• arXiv cs.CLOct 14

  Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations

  Enrico Liscio, Anne Arzberger, Aditya Shankar, Pradeep K. Murukannaiah

  Human values shape how we expect language models to behave. Dominant alignment approaches treat values as global objectives, although the appropriate value orientation is personal and contextual. Training-free alignment can steer models toward target values at inference time, yet has been largely evaluated in single-turn settings and with a single steering direction (e.g., toward safety). We evaluate four prompt- and activation-based steering methods in multi-turn conversations, where user and assistant models push toward the same or opposing, equally desirable values. We track alignment with a linear probe across three models and two value trade-offs. When assistant and user pull in opposite directions, the side matching the assistant's baseline bias often prevails, and effects that look successful early often fade or reverse after a few turns. Steering should therefore be evaluated over full conversations and in both directions of a value trade-off.

  PDF

  Summarize this
• arXiv cs.LGOct 14

  When Does Depth Help? Understanding Layerwise Inference in Tabular Foundation Models

  Wenwen Gong, Haotian Gu, Zhen Yang, Jiaying Ma, Hanqing Zhang +4

  Transformer-based tabular foundation models (TFMs) achieve state-of-the-art performance on small to medium-sized tabular prediction tasks while processing diverse inputs through a shared stack of layers. Recent layerwise analyses identify strong predictive performance in early TFM representations and substantial overlap in computation across layers. However, early predictive accessibility alone does not establish whether later layers are dispensable for every input, and aggregate accuracy can obscure which inputs benefit from additional computation. We study all 24 in-context-learning layers of TabPFN-3 across 272 classification datasets, combining representation geometry, prediction trajectories, truncation, and structural interventions, while also examining related representation dynamics in four additional TFMs. We uncover a pronounced expansion-compression trajectory, with stronger late dimensional contraction associated with larger increases in class-prototype separation. A targeted analysis of early-stabilizing samples shows that substantial representational changes persist after predicted classes stabilize, while locally ambiguous queries obtain larger accuracy gains from…

  PDF

  Summarize this
• arXiv cs.LGOct 14

  VaryST: Generalizable Spatio-Temporal Forecasting Beyond the Closed-World Setting

  Shengnan Guo, Jinjun Cheng, Xiaowei Mao, Jianhao Zhang, Haochen Lv +5

  Spatio-temporal forecasting is fundamental to many real-world systems, yet deployment environments are inherently varying: temporal distributions may evolve, spatial domains may change, and observation availability may shift from training conditions. Existing models are often tightly coupled to their training environments through node-specific representations, predefined spatial support sets, and reliance on consistently available observations, limiting generalization to changing environments. To this end, we propose VaryST, a compact spatio-temporal forecasting model trained from scratch that achieves strong generalization by dynamically constructing predictive context from available long-term observations. To accommodate temporal changes, VaryST derives node representations from long-term histories, capturing current behavior in the context of longer-term patterns. To adapt to spatial changes, it dynamically selects functionally relevant support nodes beyond predefined neighborhoods and integrates their information with recent and long-term target states and global context. To handle varying observation availability, VaryST selectively masks historical patches on which the…

  PDF

  Summarize this
• arXiv stat.MLOct 14

  Calibration and Risk-Budgeted Decisions with System One Models

  Soundouss Messoudi, Youness Mansar

  LLM workflows and agents rely on a set of binary decisions like allow/block an input, or discard/retain a retrieved document. When human deferral is unavailable, uncertain cases must still receive an action. We study System One decision scores in this setting by defaulting to a conservative fallback when uncertain, and controlling the false omission risk within the subset of inputs assigned the utility-maximizing action. Across six datasets covering moderation, safety, and reranking, we compare general System One models with task baselines and a local open-weights fine-tuned System One model. We show that naive baselines for threshold selection exceed the target risk in roughly half of settings with defined test risk, compared with 7.8\% for Learn Then Test (LTT), which leverages finite-sample statistical guarantees. Our results show why relying on a probability-like interface is insufficient for deployment decisions and how risk-budgeted selection trades useful action against observed error.

  PDF

  Summarize this
• arXiv cs.AIOct 14

  ReproBAIT: Can coding agents build their own biological AI tools?

  Shiv Muthupandiyan, Tolga H. Dur, Mac Walker, Timo Flesch, Mahmoud Ghanem +6

  Large language models are increasingly competent at leveraging scientific knowledge and existing computational tools to execute complex workflows. At present, little information exists on their ability to reconstruct tools when given limited descriptions of such tools. We introduce an evaluation titled ReproBAIT (Reproducing Biological AI Tools), which measures the ability of coding agents to design, implement, and train biological AI tools taken from published research across a range of information access conditions. The coding agent is securely sandboxed without internet access and receives training data, a fixed time and compute budget, and limited information about the tool to replicate. We then score the reproduced biological AI tool based on its performance (tool-specific metrics) on held-out data. We evaluated a set of biological AI tools of varying complexities: ProteinMPNN, an inverse protein folding tool, RFdiffusion3, an all-atom generative diffusion tool, and a closed-source molecular design tool. We institute a methodology to check for memorisation, as the open source tools are likely to be included in models' pretraining corpus. Across a range of leading closed- and…

  PDF

  Summarize this
• arXiv cs.LGOct 14

  Efficient contextual partial monitoring with general function approximation

  Kapilan Balagopalan, Hao Qin, Kwang-Sung Jun, Chicheng Zhang

  We study finite stochastic partial monitoring (PM) games with side information, a paradigm of sequential learning with partial feedback. On each round $t \in [T]$, both player and environment observe a context and player chooses an action and environment simultaneously chooses an outcome. The player neither observes the outcome nor the loss; instead, they only observe a feedback symbol which is partially informative about the outcome. We utilize the Estimation-to-Decisions (E2D) meta-algorithm of Foster et al.(2021) to design E2D.PMSide, a computationally efficient framework for contextual partial monitoring, which uses an online regression oracle to efficiently estimate the outcome distributions based on historical (context, action, feedback symbol) tuples. Since the exploration rule given by the original E2D algorithm is intractable to compute, we propose two computationally efficient exploration rules that achieve optimal regret orders in $T$: in the globally observable setting, our exploration rule is based on $ε$-greedy, while in the locally observable setting, our exploration rule uses inverse gap weighting (Foster & Rakhlin, 2020), which was previously only known to be…

  PDF

  Summarize this
• arXiv cs.CVOct 14

  Improving Category-Aware Active Domain Adaptation via Uncertainty-Guided Region Coverage

  My Tra Nguyen Huynh, Viet Cuong Ta

  Active domain adaptation researches focus on alleviating the annotation cost, which comprises source data and target data labelling, as well as enhancing the performance of unsupervised domain adaptation. However, optimizing the adaptation model based on the average accuracy can lead to the drop in performance of some difficult categories. To overcome this issue, a point-wise selection strategy built upon the influence function can be used to improve domain adaptation performance. In our work, we first address the sensitive and redundant aspects of point-wise selection, which can lead to overfitting of the acquisition function. We thus introduce a region-based approach that ensures the improvement of the adaptation model by maximizing a positive manifold coverage. An uncertainty score is further introduced to the selection mechanism to learn a better decision boundary for the target domain. In comprehensive experiments, our proposed method surpasses the first category-aware active domain adaptation baseline as well as existing active domain adaptation methods across multiple domain adaptation benchmarks. Our implementation is available at: https://github.com/nhtm2702/WHO IS NEXT.

  PDF

  Summarize this
• arXiv cs.LGOct 14

  Asymptotically Exact Negative Guidance of Diffusion Models via Positive-Unlabeled Learning

  Hirohane Takagi, Masashi Sugiyama

  Negative guidance of diffusion models generates samples that avoid an undesired condition, specified either by reference samples or by a conditional score alongside an unconditional model. Existing methods inject repulsive terms derived from local information at each sample, so the resulting sampler does not target an explicit density, and its bias cannot be systematically reduced by increasing computation. We instead present a negative-guidance method that samples from an explicit target density, which uniformly handles both reference-sample and score-based specifications and suppresses probability mass near the undesired condition with an avoidance strength that parametrizes the target density itself. Realizing this target requires estimating, at each diffusion timestep, how likely each sample is to belong to the undesired condition. We cast this estimation as a Positive-Unlabeled learning problem that admits lightweight online training along the diffusion trajectory, using condition samples as positives and diffusion particles as unlabeled data. Because the guided particles drift from the unconditional marginal, we further correct the residual mismatch via Sequential Monte…

  PDF

  Summarize this
• arXiv cs.CLOct 14

  Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs

  Zhenyu Zhang, Zhichao Cao

  An agent that drives an LLM sends the model the same conversation again and again, and each turn adds only a little text at the end. Inference engines reuse the KV cache of the shared prefix, so only the new tokens pass through the model. However, tokenization has no such shortcut. The CPU front end of the serving engine converts every request back into token IDs from its first character, which lengthens the time to first token (TTFT) and consumes CPU on every turn. We present Delta Tokenizer, a CPU tokenization front end that tokenizes only the delta between a request and the previous request of the same session. It retains the earlier token sequence, encodes again only the appended text and the short tail whose tokens the append could affect, and joins the two only at a position the tokenizer's splitting rules themselves cannot move, which keeps the result exact. For long requests with no state to reuse, it cuts the text where practical tokenizers already keep neighboring pieces independent, encodes the pieces on several cores at once, and concatenates the outputs into the reference encoding, exactly. A shared CPU pool schedules these chunks together with the short incremental…

  PDF

  Summarize this
• arXiv cs.CLOct 14

  Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems

  Seojin Hwang, Youngmin Choi, Hwanhee Lee

  Recent work lets LLM agents evolve their own harness, up to the team itself. Yet in human groups, the same abilities achieve different outcomes under different social conditions. Borrowing the notion of scaffolding from developmental psychology, we investigate whether the social scaffolding of a self-evolving multi-agent system (MAS), namely its team composition and structure of interdependence (e.g., collaboration, conformity, or competition), shapes how the team evolves. We present AGORA (Agents Growing through Organization and Relational Arrangements), a framework that varies the two axes independently across three evolution mechanisms. Across web search, codebase understanding, and document QA, we find that (1) evolution alone rarely helps, while the right scaffolding surpasses a strong agent evolved alone by up to 25 points; (2) it does so by moving authority to the most capable member while the others step back; and (3) this member, having grown within the team, outperforms the same model alone. Designing a self-evolving MAS thus requires designing its social scaffolding, which shapes not only how its agents learn, but who gets to lead.

  PDF

  Summarize this
• arXiv cs.AIOct 14

  Learning Perception to Action Rules through Reinforcement Learning for Theory-of-Mind Reasoning

  Soyoung Oh, Nursulu Sagimbayeva, Nicholas Jennings, Paulina Hussein, Mehwish Alam +1

  In Theory-of-Mind (ToM) reasoning, large language models may correctly infer the agent's mental state but fail to use that inference when predicting or judging the agent's subsequent actions. In this work, we investigate whether explicitly connecting perception, belief, and action with rules improves ToM reasoning. We generate concise \textit{if--then} rules that express relations between perception to action and use Group Relative Policy Optimization (GRPO) to train Qwen2.5 and Llama3 models across four ToM benchmarks. GRPO with rule generation outperforms prompting, supervised fine-tuning, and think reasoning. Moreover, rules generated by larger models ($\geq$3B) provide transferable knowledge that improves downstream performance of GPT-4o-mini. These findings suggest that explicit rules help models bridge the reasoning gap in ToM and support knowledge transfer across models.

  PDF

  Summarize this
• arXiv cs.CVOct 14

  StateCraft: Augmenting Video World Models with Abstract-State Reasoning

  Lorenzo Venturoli, Nedko Savov, Xi Wang, Marc Pollefeys, Luc Van Gool

  Video world models aim to simulate complex environments with realistic appearance and dynamics. Maintaining consistency over long interactions requires tracking past actions and understanding how those actions affect the environment. Video world models can only generate partial observations of the entire world at a time. Remembering past observations alone may not be sufficient in every scenario, as they may no longer reflect the world after a sequence of actions. We introduce a paradigm that decouples a persistent, updatable abstract state from the visual state. The abstract state keeps track of the entities' changes as the world evolves, and is used to condition the visual state, which represents the current view. We propose StateCraft, a novel architecture that pairs a language model with a video diffusion model to evolve the dynamic world. We evaluate our approach on generative interfaces, which pose a substantial challenge because a single action may change parts of the state that are not currently visible. To support training and evaluation, we develop a data collection pipeline that produces informative, densely annotated trajectories at scale without human or agent…

  PDF

  Summarize this
• arXiv cs.CLOct 14

  JetSpec++: Foresight Parallel Drafting with Self Refinements

  Zhaoxiang Feng, Lanxiang Hu, Haoran Yuan, Yu-Yang Qian, Jihui Yang +4

  Speculative decoding's effectiveness is governed by two quantities: the acceptance rate of drafted tokens and the cost of producing them. We show the two objectives can be co-optimized with algorithm-system co-design, and present JetSpec++, which retains a strong learned draft model while hiding its cost almost entirely. JetSpec++ combines three components: intra-block self refinement, which trains the draft model to iteratively correct its own parallel predictions within a block of tokens; inter-block foresight drafting, which begins the next block before the current one is verified; and pipelined drafting and verification, which overlaps the two stages so drafting's marginal wall-clock cost approaches zero. In comparison with state-of-the-art speculative decoding methods, JetSpec++ attains a wall-clock speedup of up to $7.45\times$ over autoregressive (AR) decoding and $1.54\times$ over DSpark, with up to 1,361 tokens per second for a single request on GB200, and an average acceptance length of 8.94 across seven benchmarks, while preserving the target model's output distribution. Our code and models are available at https://github.com/hao-ai-lab/JetSpec.

  PDF

  Summarize this
• arXiv cs.MAOct 14

  One Agent per Zone: Decentralised Active Inference under Partial Observability for Multi-Zone HVAC Control on Edge Hardware

  Panagiotis Kasnesis, Lazaros Toumanidis, Christos Chatzigeorgiou, Amalia Contiero Syropoulou, Miguel De Prado

  Multi-zone HVAC control is a shared-resource problem: zones on a floor share one fan and cooling coil, so each zone's demand affects its neighbours and the building's peak power. One controller per zone scales well and isolates faults, but each controller then observes only its own zone, not the shared air loop. In this work, we present a decentralised controller that assigns each zone an independent active-inference agent, which plans by minimising expected free energy and learns its zone dynamics online. The agents exchange no messages, so none depends on peers or a coordinator; each accounts for the unobserved coupling through a latent disturbance and a congestion signal derived from the metered power of the other floors. Fifteen agents are evaluated in EnergyPlus on a Sinergym office building against three rule-based controllers and three MADDPG fleets, one given the same floor meters. On five noisy weather years unseen during training, the fleet attains the lowest mean building peak (88.3 kW, 3 to 17 % below the baselines) and staggers the floor peaks better than any baseline, with a coincidence factor (building peak over the sum of the floor peaks) of 0.929 versus 0.936 to…

  PDF

  Summarize this
• arXiv cs.CLOct 14▲ 82

  More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning

  Linhao Zhong, Bozhen Fang, Xuehao Wang, Duochao Shi, Zekai Luo +5

  Self-distillation in agentic reinforcement learning has produced a growing number of methods. However, existing evaluations often reuse published baseline scores under differing experimental conditions, provide few direct comparisons among contemporaneous methods, and give limited attention to training stability and computational cost. These limitations obscure the source of reported gains and the practical strengths and weaknesses of competing methods. Costly agent training further motivates a common basis for comparison. To address these gaps, we compare ten representative methods against GRPO on ALFWorld, ScienceWorld, Search-QA, and WebShop across three model sizes. Our study comprises 132 training runs totaling 12,124.4 recorded H800 GPU-hours, with repeated evaluations over eight rollout seeds for the pretrained backbones and all eleven trained methods. Beyond final task performance, we examine training progress, stability, and measured computational cost. Our evaluation yields three key findings. First, gains over GRPO remain limited in scope, with no method consistently outperforming it across settings. Second, strong final performance can mask unstable training. Third,…

  Code

• Adversarial training47−4.1%
• Federated learning42−4.5%
• Forecasting time72−8.9%
• Frontier LLMs425−16.7%
• Inference & serving241−19.7%
• 3D scene144−20.0%
• Agents470−20.1%
• Anomaly detectors30−21.1%