Research

• General in-context learning for embodied agents1 paper raises it · 0 address it

  Improving adaptability of models to new tasks and environments without task-specific fine-tuning is critical for scalable embodied intelligence.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Cost and latency in agentic systems2 papers raise it · 1 addresses it

  High computational and time costs limit practical deployment of agentic systems for real-world tasks.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
  • PremiseCheck: Efficient Fact-Checking of Human and AI Peer Reviewsraises · addresses
• Physical-state verification and failure recovery1 paper raises it · 0 address it

  Robustness to irreversible failures and accurate state tracking are essential for reliable embodied control.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Evidence-grounded multimodal reasoning benchmarks1 paper raises it · 1 addresses it

  Current benchmarks often fail to ensure models use necessary audio-visual evidence, weakening evaluation reliability.

  • EvaOmni: An Evidence-Grounded Benchmark for Audio-Visual Reasoningraises · addresses
• Length generalization in sequential state tracking1 paper raises it · 1 addresses it

  Models struggle to maintain performance on problems much longer than their training data, limiting scalability.

  • PAR: Progress-Aligned Recurrence for Length Generalization in State Trackingraises · addresses

Named by an AI from these papers' abstracts.

• Training & fine-tuning42+90.9%
• AI safety29+61.1%
• Computer vision29+31.8%
• Federated learning47+27.0%
• Evals60+25.0%
• Code models13+18.2%
• Quantum machine learning24+14.3%
• Image & video generation20+5.3%

• arXiv cs.AIOct 16

  Reasoning Models Can Control Their Chains of Thought When They Tell Themselves To

  Uriel Dolev, Idan Schwartz, Yoav Goldberg

  Chain-of-thought (CoT) controllability, the ability of reasoning models to control properties of their CoT, is consequential for both their usefulness and reliable oversight. Recent work evaluates this capability by measuring whether models follow CoT constraints given in the user prompt, and reports very low controllability across models. We argue that this low controllability may not reflect an inherent inability of reasoning models to control their CoT, but instead result from naively specifying constraints in the user prompt. To test this, we introduce Self-Instructed CoT Controllability: a complementary evaluation that places the constraint as a self-instruction inside the model's CoT and measures its ability to follow it. We show that, under self-instruction, models achieve significantly higher controllability than with standard user instructions. Even a generic reminder to follow the user's constraint substantially improves controllability. Moreover, effective self-instructions can be bootstrapped into the reasoning through a simple instruction in the user prompt, enabling improved control even without access to the reasoning channel. Finally, using self-instruction, we…

  PDF
• arXiv cs.AIOct 16

  Beyond Reward Scoring: Online Multi-Agent Preference Learning for Multimodal Response Generation

  Ziyun Zou, Jiwoo Hong, Yifan Lu, Shao Tang, Weiyi Wu +3

  Multimodal preference alignment often follows a score-based RLHF pipeline: a reward model is first trained to learn human preferences, then used to score on-policy responses and optimize the response model's generation. This turns preference supervision into a scalar signal for generation: a model may learn to produce answers with higher preference scores, but is rarely trained explicitly to learn the reasons underlying human preferences. We therefore ask whether making preference understanding an internal capability of the response model can improve its generation. To investigate this question, we propose \textbf{PrefMind}, an online dual-agent reinforcement learning framework with two trainable agents: a response Generator and a preference Critic. Given a pair of candidate responses, the agents analyze and discuss which response better aligns with human preference. Human labels supervise their judgments and corrections. For pairs on which both the Generator and revised Critic correctly identify the human-preferred response, the Generator produces an improved answer, and the Critic compares it against the original preferred response to provide a generation reward. Both agents…

  PDF
• arXiv cs.LGOct 16

  Sample, Mix, or Verify? Allocating Test-Time Compute under Imperfect Verifiers

  Srinivasan Iyengar

  A common way to make language models and agents more accurate is to spend more compute at inference time: generate many candidate solutions and submit the one a verifier prefers. Success then requires that some candidate is correct and that the verifier picks it. The first grows predictably with the number of samples; the second does not, because real verifiers (unit tests, majority votes, LLM judges, reward models) accept wrong answers, reject right ones, and make correlated mistakes. With a fixed budget, a practitioner must decide how many samples to draw, which models to draw them from, and how much to spend on verification, and today these choices are made by rules of thumb. We introduce SPLIT, which makes them from a small labelled pilot. It rests on an exact formula for the probability of submitting a correct answer under accept/reject and top-score selection that accounts for varying task difficulty, blind spots shared across models, and correlated repeated verifier calls. SPLIT fits the formula by empirical Bayes and uses it to plan budgets on unseen tasks. Across seven benchmarks spanning math, science QA, code and software-engineering agents, 26 models, 175 agent…

  PDF
• arXiv cs.CLOct 16

  Ignored, Not Refused: What Low Prompt-Injection Success Rates Hide in Small Tool-Using Agents

  Farhan Zarif, Kazi Shahid Ahmed Galib, Synthia Akter Promi, Shabnam Sharmin Tarin, Ariyan Hossain +2

  Tool-using agents read untrusted external content, which exposes them to indirect prompt injection (IPI), where hidden instructions hijack the agent's tool calls. Benchmarks measure IPI security by the attack success rate (ASR), which treats every failed attack as a successful defense. To test this assumption, we audit 33,000 trajectories from eleven open-weight agents (2-14B parameters) on InjecAgent and AgentDojo. A label-blind LLM judge shows that most failed attacks are ignored (76-80%) rather than refused (17-18%). With identical model weights, changing only the system prompt shifts InjecAgent ASR from 0.2% to 43%. Our Attack-Task Outcome Matrix (ATOM) further shows that in 21.7% of AgentDojo episodes, the agent stayed safe but did not complete the user's task. A low ASR therefore mostly reflects agents silently ignoring injections, not refusing them. Evaluations should report how attacks fail and whether the user's task is completed, alongside ASR. We release the labeled corpus, judge prompt, and analysis code.

  PDF
• arXiv cs.MAOct 16

  LLM Evolution in a Simple Simulated Market

  Charlie Pilgrim, Christopher Krupenye, Dora Biro, Richard P. Mann

  Arguments that AI systems will evolve in harmful ways are largely made in the abstract. We provide a concrete empirical example. We simulate simple markets consisting of populations of twenty Large Language Model agents tasked with solving arithmetic problems. Agents differ only in their system prompt; those that go bankrupt are removed and replaced by mutated copies of the prompts of successful agents. Two market conditions select for opposite phenotypes: a market that rewards accuracy produces agents with high accuracy (94%) and high token use; a market that charges per token produces agents with low accuracy (10%) and low token use. In the second market, only tokens are incentivised yet we see a collapse in accuracy. Token use and accuracy are correlated traits such that selection on one also moves the other, and behaviour adapts along an unpriced dimension. A third neutral condition with no selection allows us to separate these effects from bias in the mutation operator. Adaptation occurs through a collective optimisation process that operates above the level of individual agents, who are never told the payment rule, do not receive feedback, and have no memory.

  PDF
• arXiv cs.NIOct 16

  Preventing Data Hallucinations in Agentic AI for Optical Network Troubleshooting

  Jeremias Dötterl, Martin Kuipers

  An agentic AI system for automated troubleshooting in optical communication networks is demonstrated that integrates data references into the agent architecture to prevent data hallucinations, improving the interpretability and trustworthiness of root-cause identification and reasoning.

  PDF
• arXiv cs.LGOct 16

  FLARE: Physics-Grounded Evaluation of Language-Model Agents for Feeder Restoration with Degraded Observations

  BHVSP Subrahmanyam, Neeraj Choudhary

  Restoring service after distribution-grid faults requires decisions from incomplete and conflicting information while respecting electrical constraints. Addressing this need, we introduce FLARE (Feeder-Level Agent Restoration Evaluation), a physics-grounded benchmark for post-isolation feeder restoration. FLARE con- tains 800 AC-feasible cases across five observation settings and allows agents to gather evidence, evaluate switches, and perform restoration actions under fixed operational limits. We evaluate five instruction-tuned models over 78,750 episodes, including topology, distributed-generation, and loading shifts. Mean normalized regret ranges from 0.082 to 0.148. Although every model has zero median regret, P95 regret ranges from 0.814 to 1.000, revealing an important failure tail. Shadow replay shows that 96.4% of 15,706 guard-rejected actions would have caused a physical violation. FLARE enables reproducible evaluation of language-model based restoration support while emphasizing the need for external safety checks.

  PDF
• arXiv cs.CVOct 16

  LonGaussian: Tracking Foundation Priors for Long-Context Feedforward Gaussian Reconstruction

  Zhoufeng Xie, Guanjun Wu, Shuai Zhang, Xinggang Wang, Wenyu Liu

  Reconstructing high-quality, renderable 3D scenes from long videos is essential for robotic perception and embodied intelligence. Many feed-forward Gaussian reconstruction methods rapidly predict 3D scenes from a small set of images, but extending them to longer sequences presents two bottlenecks: jointly processing all views increases GPU memory requirements, while accumulating pixel-aligned predictions redundantly represents the same surfaces and increases storage and rendering costs. We present LonGaussian, a streaming feed-forward framework that reconstructs compact, globally renderable Gaussian scenes from long image sequences with known intrinsics and unknown camera poses. Built on a pretrained streaming 3D reconstruction backbone, LonGaussian decodes long-context geometric features into dense Gaussians. Within each chunk, multi-frame point tracks link repeated observations of the same surface and guide a learned merger to consolidate the corresponding Gaussian predictions. Rendered-depth consistency filtering further suppresses Gaussians from other chunks that incorrectly occlude locally reconstructed surfaces. On DL3DV and RealEstate10K benchmarks, LonGaussian improves…

  PDF
• arXiv cs.CLOct 16

  Who Pays for Efficient Inference? Confidence-Based Small-to-Large LLM Routing Across Tasks and Languages

  Md Sabbir Hossen, Anichur Rahman, Pabon Shaha, Airin Afroj Aishi, Andrew H. Sung +1

  Small-to-large language model routing can reduce inference cost by escalating difficult inputs while retaining easier ones on a smaller model. Its success, however, depends on identifying examples that benefit from escalation, and the reliability of confidence across multilingual tasks remains unclear. We present a systematic evaluation of confidence-based routing across three model families, four languages, and three task types, covering 72 model, dataset, language, and scale configurations and 42,288 predictions. We compare global confidence routing with random routing, equal-quota confidence routing, and an oracle upper bound. Confidence identifies beneficial escalations on the Belebele reading comprehension and SIB-200 topic classification datasets, with mean correction-detection AUCs of 0.612 and 0.705. At 50% escalation, it improves over random routing by 1.40--4.67 percentage points on Belebele and 2.54--5.63 points on SIB-200. In contrast, the mean AUC on the MMLU-ProX-Lite knowledge and reasoning dataset is only 0.510, and routing provides no consistent improvement over random selection because small models often remain highly confident while incorrect. Global confidence…

  PDF
• arXiv cs.CVOct 16

  Can the Parts Fool the Test? Counterfactual Pairing Cycles for Relational OOD

  Yibo Zhou, Yirong Xiang

  Relational out-of-distribution errors are easy to describe and hard to evaluate cleanly. An image may be natural, a caption may be fluent, and all mentioned objects may be common, yet the caption may describe the wrong relation in the image. Many evaluations intended to test this behavior can be solved without checking the pairing at all, because negatives differ in caption style, template, length, object frequency, or image statistics. We propose counterfactual pairing cycles, a blocked evaluation primitive that reuses the same two images and the same two captions on both sides of the comparison, changing only which image is paired with which caption. This makes image-only, text-only, and additive component shortcuts cancel exactly in finite samples. The same contrast also gives a training loss for relational outlier exposure and a diagnostic for bilinear vision-language scores. Across synthetic controls, GQA, and COCO, cycles separate coarse image-caption matching from directional role-binding failures, while also identifying geometric size relations that remain strongly visible in global features. In a shortcut-contaminated COCO outlier-exposure stress test, a model selected…

  PDF
• arXiv cs.CLOct 16

  Think More or Read More? A Token-Budget Theory of Reasoning with Shared Context

  Manh Nguyen, Sunil Gupta, Hung Le

  Reasoning systems increasingly improve answers by incorporating additional context into the model's input, from multi-agent debate, where agents read their peers' responses, to self-refinement, where a model revisits its own drafts. In these systems, communication consumes context as one agent's output becomes another's input. We study the regime where token reading and generation draw on one fixed token window, so every token spent reading is a token that cannot be spent reasoning. We develop a \textbf{token-budget theory} that formalizes this trade-off, modeling accuracy as a function of the reasoning budget remaining after reading and aggregating votes with a Condorcet jury model that permits correlated errors. For multi-agent debate, the theory gives a closed-form prediction for the optimal number of agents. This quantity is estimated from single-agent measurements, allowing the optimal group size to be determined without running debate, and the same accounting extends to a heterogeneous set of different models. We empirically verify our theory and show that, at the predicted group size, our method outperforms strong baselines under the tightest calibration spend, where…

  PDF
• arXiv cs.CLOct 16

  MGE-PO: Entropy-Aware Credit Assignment for Mathematical Reasoning

  Jihao Huang, Guangjia Chai, Zhirui Chen, Lei Wang, Yao Liu +7

  GRPO assigns the same group-relative advantage to equally rewarded responses within a question, leaving their generation uncertainty unused. We introduce MGE-PO, centered on four-quadrant (4Q) advantage shaping that combines verified correctness with rollout generation entropy. Correctness determines the sign of an additional correction, while entropy controls its magnitude: lower-entropy successes receive larger bonuses and higher-entropy failures receive larger penalties. An online entropy-based question curriculum supports this credit rule using cached rollout measurements. At the final checkpoint after eight training epochs on Qwen3-8B-Base, adding 4Q across three policy optimization settings yields absolute gains of 3.23-3.69% in six-benchmark macro avg@32 and 4.24-11.06% in macro pass@32. On Qwen2.5-7B-Base, the complete configuration raises four-benchmark macro avg@32 from 41.5% to 45.4%.

  PDF
• arXiv cs.LGOct 16

  Anisotropic Representation Dynamics in Transformers

  Nguyen Do, My T. Thai

  Small differences between internal representations can shrink or grow as they pass through a Transformer layer, affecting how information is preserved and how sensitive the model is to perturbations. Yet it remains unclear when changes in attention weights can turn a representation difference that would otherwise shrink into one that grows. We develop a \textbf{Mixture-Gated Theory of Anisotropic Representation Dynamics} that separates self-attention into value transport, query gating, and key gating. Our analysis reveals how the geometry of attended representations constrains gating and how interactions between these mechanisms can reverse contraction induced by value transport. We formalize this phenomenon as gating-induced regime switching, where a representation difference contracts under fixed attention routing but expands under full attention. We derive a necessary and sufficient spectral criterion that certifies the existence or absence of switching and identifies switching directions without exhaustive search. Beyond existence, our theory characterizes maximum relative switching strength, certifies the severity of contraction that gating can overcome, and determines the…

  PDF
• arXiv cs.SEOct 16

  Towards Agentic Proof Harness Engineering

  Rohith Kumar Shanmuganathan, Cedric Richter, Heike Wehrheim

  Formal verification of real-world projects requires substantial manual effort, from specifying correctness properties to constructing the infrastructure needed to check them. We envision verification agents that assist developers throughout this process. Among the many tasks involved, one stands out as both foundational and amenable to automation: proof-harness engineering. A proof harness defines an isolated verification environment for a program unit, such as a function or module, by providing symbolic inputs, initialization code, and models of its dependencies. Today, proof-harness engineering remains largely a manual task performed by verification engineers. To explore the potential of coding agents for this task, we propose an agentic proof-harness engineer that automatically constructs verifier-ready harnesses from a software project and a designated verification target. We combine the agent with formal soundness and precision guards to validate generated abstractions and guide their refinement. We evaluate the feasibility of our approach in an exploratory case study on curl and libuv, using three historical assertion violations already embedded in the source code as…

  PDF
• arXiv cs.LGOct 16

  Response-Informed Value Estimation for Cost-Aware Multi-Turn LLM Routing

  Haotian Yan, Jiaqi Liu, Bingyi Jing, Tianjun Yao

  Multi-turn LLM routing selects a model at every turn before its response is available, yet task success is revealed only when the interaction ends, leaving each intermediate choice without a label of its own. Logged trajectories, however, retain the response each call produced, which exposes what each turn required and how well the chosen model served it, evidence that outcomes alone hide. We introduce RIVET (Response-Informed Value Estimation for Turn-Level Routing), which exploits this post-call evidence during training while keeping routing strictly pre-call at deployment. RIVET jointly trains two value estimators over a shared state-model representation: a deployable Routing Value Estimator that scores candidate models from pre-call information, and a training-only Response-Informed Value Estimator that additionally observes the realized response. Conditioning on the response turns response-induced return variation from residual noise into an observed covariate, and we show that this reduces the leading asymptotic estimation error of the shared representation. On the MINT benchmark, RIVET solves more held-out tasks than the strongest multi-turn baseline while using 74.7% less…

  PDF
• arXiv cs.CVOct 16

  Kingfisher: Fast Personalized Visual Autoregressive Generation via Scale-Aware Prompt-Conditioned Scale Decomposition

  Dinh-Khoi Vo, Akihiro Sugimoto, Minh-Triet Tran, Trung-Nghia Le

  Personalized image generation has been extensively studied in diffusion models, often relying on subject-specific optimization of model weights, textual embeddings, LoRA modules, or adapters. In this work, we explore a different direction for fast personalization in visual autoregressive models. Our key observation is that a short prefix of early autoregressive scale codes is sufficient to strongly determine both object structure and global appearance, while later scales primarily refine the established generation trajectory. Based on this insight, we introduce Kingfisher, a model-tuning-free framework that decomposes early residual scale codes into content- and style-related components through prompt-conditioned scale-code subspaces. Given two real images, Kingfisher first decomposes their early scale codes, then recomposes the content components of one image with the style components of the other into a compact semantic prefix. The frozen generator subsequently predicts the remaining scales through bitwise self-correction for coherent refinement. Experiments on CSD-100 show that Kingfisher achieves strong content preservation and the highest style and text alignment among the…

  PDF
• arXiv cs.CVOct 16

  X-Splat: Unified Gaussian Splatting for Joint RGB and X-ray 3D Reconstruction

  Wangze Xu, Chu Zhou, Mae Yamaguchi, Linrui Dai, I-Chao Shen +2

  Jointly reconstructing RGB and X-ray provides new insights by connecting external appearance with internal structure, enabling a more complete understanding of real-world objects than single modality. Yet existing methods train separate models per modality and suffer from redundant geometry, alignment artifacts, and loose coupling that limits cross-modal insight discovery. We propose a unified representation in which all Gaussian primitives share a single set of geometric parameters, with only surface-proximal primitives additionally carrying appearance attributes. We perform cross-modal registration through clean multi-view stereo and X-ray geometric proxies instead of aligning two sets of optimized Gaussians, improving alignment stability. We then jointly optimize the unified representation via supervision from both modalities, complemented by a surface-depth consistency loss derived from X-ray geometry to guide RGB reconstruction. Experiments on the InsideOut dataset and our Flower & Insect dataset show that our framework achieves more robust cross-modal alignment, higher RGB rendering quality without compromising X-ray reconstruction, and a more compact representation than…

  PDF
• arXiv cs.CLOct 16

  MultiAgent-OPD: End-to-End Training of Hierarchical Multi-Agent Systems with On-Policy Distillation

  Yuhan Sun, Jiaxin Zhao, Yongdong Luo, Yibo Hu, Meiguang Jin +4

  In a hierarchical multi-agent system, a Leader assigns tasks and subagents carry them out. Jointly training the Leader and the subagents is difficult because a final answer gives only indirect feedback to the subagents. An LLM judge can score the performance of the subagents, but these scores depend on the judge's own assessment of the quality of subagents' responses. We present MultiAgent-OPD, which combines on-policy distillation with reinforcement learning to jointly train a shared-parameter Leader-Searcher-Answerer team end to end. We first train a Teacher in a Searcher-Answerer workflow using answer correctness. During team training, the frozen Teacher scores student-generated search and summary tokens to provide local process supervision for the Searchers, while the Leader learns from team success and the Answerer from candidate correctness. Across seven datasets, the trained system reaches 38.68% macro avg@4, exceeding Prompted Critic by 4.06 percentage points and C-MAS + RL, the trained cascaded multiagent system, by 2.61 percentage points. The team also solves more questions within four attempts.

  PDF
• arXiv cs.CVOct 16▲ 34

  RAP: A Dataset for Paper-to-Poster Generation via Declarative Markup from Human-Crafted Exemplars

  Haonan Han, Yukang Lin, Minliang Lin, Haoxian Tan, Jun-Yan He +5

  Making a conference poster is visual coding in miniature: an entire paper must be compressed onto a single page, which exercises summarization, visual design, and code implementation at once, using nothing beyond the paper itself. Yet benchmarks for this task in the context of visual coding remain scarce: existing resources are small and annotated only with layout boxes, so evaluation must fall back on VLM-as-judge or human checklists rather than objective render-and-compare, and the generation methods they support design layouts from scratch, leaving human-crafted posters unusable as either a retrievable design library or as verified supervision. We introduce RAP (Reconstruct A Poster), a dataset and benchmark for paper-to-poster generation as a visual coding task. RAP contains 25K+ verified paper-poster pairs from various machine learning conferences held between 2022 and 2026, annotated with a three-level content taxonomy, plus a held-out benchmark of 150 test papers each paired with its human poster. A fully automated derendering pipeline turns each poster into styled HTML/CSS of typed, refillable slots, so every sample is a paper-poster-markup triplet verifiable by rendering…

  Code
• arXiv cs.AIOct 16

  Characterising and Analysing Retrieval-Augmented Generation Techniques

  Emmanuel Igbekele, Olawande Daramola

  The advent of Large Language Models (LLMs) across many application domains has increased the need for Retrieval-Augmented Generation (RAG) to reduce hallucinations and improve LLM performance. Thus far, several RAG techniques with impressive features and capabilities have been proposed in the literature, making it a herculean task to select the most appropriate RAG techniques for specific problem scenarios. Thus, we propose a conceptual framework comprising main dimensions and sub-dimensions for characterising (defining "what it is") and analysing (defining "how well it does") existing RAG techniques. The conceptual framework was developed through a qualitative analysis of the attributes of RAG techniques reported in the literature. It can serve as a pre-empirical filtering mechanism for researchers and practitioners in selecting the highest-fit RAG candidates that meet specific user and task requirements before empirical benchmarking of RAG techniques, which can be computationally and financially expensive.

  PDF