跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1072新增论文
+845其中 AI 相关
+94已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-09 21:43 · 全站最近入库：2026-11-09 13:22各信源进度 →

统计口径与数据新鲜度

统计 2026-11-08 21:53 — 2026-11-09 21:53（北京时间） · 点击数字查看对应简报；AI 相关包含在新增论文内。

最近完成检查：2026-11-09 21:43最近发现新增版本：2026-11-09 13:22库内论文最新更新时间：2026-11-07 02:59

每分钟刷新已保存的数据。空检查不代表采集异常；新收录可能包含历史论文。AI 导读按新增版本计数。

18,172 已收录论文15,506 AI 相关4,527 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →

找到 15,506 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-09 21:40（北京时间）
新论文 · v1arXiv:2611.02666

Train the Sampler You Run: Few-Step Masked Diffusion Language Models via Mixture-Likelihood Fine-Tuning

Amin Najafgholizadeh

原文摘要

When a masked diffusion language model is sampled in a few parallel steps in a random order, the model actually run is a mixture over reveal plans, while cross-entropy training fits token conditionals rather than this mixture. We introduce mixture-likelihood fine-tuning (Mix-FT), which fine-tunes a pretrained model on an importance-weighted bound of its own sampler's likelihood, with no teacher and no change to the architecture or the sampler. On LLaDA-8B, a LoRA fine-tune for 0.002% of the pretraining compute writes better OpenWebText continuations than continued cross-entropy and Alpha diffusion given the same compute, at every tested step count. With half the model evaluations, 16 rather than 32 steps per block, its continuations score better than the original model's in every seed under three judges, with confidence-based decoding as well as random order, and under confidence-based decoding two blinded LLM raters from different model families prefer them in 69 and 71% of pairs. On a public 110M-parameter block diffusion model, Mix-FT takes under four T4-hours and beats continued cross-entropy, Alpha diffusion and SDTT distillation at 4, 8 and 16 steps per block, under two judges in every seed. Exact likelihoods on text8, with a control that weights every plan equally, attribute about three quarters of the likelihood gain over continued cross-entropy to weighting each plan by how well it explains the sequence.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.20290

PhysOmni: Physics-Grounded Multi-Object Scene Generation from a Single Image with Real-Time Interaction

Xin Zhang, Yabo Chen, Yijie Fang, Wanying Qu, Haibin Huang, Chi Zhang, Feng Xu, Xuelong Li

原文摘要

Recent generative video models achieve impressive visual quality but remain constrained by limited physical consistency and controllability. Existing video generation methods provide minimal physical control, and single-image-to-3D conversion approaches often suffer from object interpenetration. Furthermore, physics-based scene-level 3D generation methods exhibit spatial misalignment, stylized artifacts, and inconsistencies with the input data, restricting their use in realistic interactive video synthesis. We propose PhysOmni, a training-free framework that converts a single image into a physically consistent and controllable video through holistic scene-level 3D reconstruction. By rep?resenting the full scene geometry in a unified spatial coordinate system, PhysOmni resolves object penetration and alignment ambiguity. Unlike prior methods, this formulation enables accurate scene?level multi-object interactions and introduces richer, complex control types for advanced mechanics?based manipulation. By decoupling simulation from rendering, PhysOmni bypasses latency-heavy priors, achieving real-time physical interaction previews paired while preserving photorealistic visual fidelity. Experimental results demonstrate that PhysOmni substantially outperforms prior methods in physical fidelity, spatial coherence, and controllability. Project Page: https://zephipage.github.io/PhysOmni/

cs.GR · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02661

PriCBMs: Multi-granular Concept Bottleneck Vision-Language Models for Visual Privacy Recognition

Shenglei Pei, Haotian Sun, Pengfei Li, Bingxu Lu, Wenlong Yu

原文摘要

Widespread image sharing on social media has increased the need for visual privacy recognition to identify inadvertently exposed privacy information. However, multi-label visual privacy recognition presents two challenges. First, privacy categories represent high-level semantic attributes and do not have a one-to-one correspondence with visual objects. Different visual objects may involve the same privacy category, while the same type of object may contain different types of privacy-sensitive information. In this setting, direct prediction methods generally lack explicit guidance for identifying category-relevant privacy cues, leaving this process implicit in the learned image-to-label mapping and potentially limiting classification performance. Second, privacy cues are often distributed across image regions and may not be adequately represented by whole-image features. To address these challenges, we propose Privacy Concept Bottleneck Models (PriCBMs). By introducing privacy-related concepts as intermediate semantic representations, PriCBMs first identify category-relevant privacy cues through visual-concept matching and then predict privacy categories based on concept responses. This process makes privacy predictions more interpretable. To improve the quality of these concepts, we design a concept selection strategy based on target-category relevance, cross-category discriminability, and semantic diversity. Meanwhile, we introduce a global-region modeling module to capture localized privacy evidence while retaining global context. Experiments on VISPR and FineViP show that PriCBMs outperform evaluated competitors across all four multi-label metrics. Ablation studies and concept-quality analyses support the effectiveness of the proposed design. Code is available at https://github.com/sun2293674491-dotcom/PriCBMs.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2604.05484

CoEnv: Code-Driven Exploration in Real-to-Sim Digital Twins for Multi-Arm Manipulation

Li Kang, Yutao Fan, Rui Li, Heng Zhou, Yiran Qin, Zhemeng Zhang, Songtao Huang, Xiufeng Song 等 14 位作者

原文摘要

General-purpose coding agents can now write, debug, and refine entire robot programs, but only if they have somewhere to practice. Physical hardware is a poor practice ground: exploratory rollouts are slow to reset and, when several arms share a workspace, unsafe. Generic simulation benchmarks are safe but do not match the geometry, assets, or viewpoints of the workspace where the robot must actually work. We present CoEx (Code-driven Exploration), which lets the agent practice in an executable digital twin of its own deployment workspace. The twin is built by CoEnv, our real-to-sim framework, which reconstructs the physical scene from multi-view RGBD into a physics simulation. Inside it, a centralized code agent explores at the program level, writing a complete multi-arm program, running it in the twin, and refining it against grounded verification (privileged state checks, checkpoint predicates on risky steps, and adaptive viewpoints against inter-arm occlusion). Verified programs transfer to hardware through collision-checked deployment. On five real-world multi-arm tasks with up to three heterogeneous robots, CoEx reaches 70% average success, against 50% for the same system without the code agent (where a VLM must choose every primitive step by step), outperforms two prior VLM-planning frameworks, and produces verified episodes at 3.8-11.7x the ablated rate. Because every verified rollout is also a demonstration, exploration leaves reusable experience behind: a $π_{0.5}$ policy fine-tuned purely on 200 agent-generated rollouts reaches 100% and 70% real-robot success on two tasks, with no human demonstrations.

cs.RO · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02660

HeLM: Hierarchical Explicit Language Memory for Non-Markovian Manipulation

Gyeonghun Kim, Donghoon Kim, Minsoo Choi, Suyun Lee

原文摘要

Recent advances in Vision-Language-Action (VLA) models have enabled fine manipulation capabilities, yet performing tasks that require working memory remains a key challenge. Most existing VLA architectures operate under a Markovian assumption, mapping only the current observation to an action, which leads to failure in non-Markovian environments. Inspired by the human 'inner monologue', we propose Hierarchical Explicit Language Memory (HeLM), a hierarchical VLA framework that maintains task-relevant history in explicit language through factorized event detection and event-triggered memory updates. We evaluate HeLM across 10 non-Markovian real-world tasks with 56 variants, including five newly formulated inter-episode tasks that require persistent context across task boundaries. HeLM improves intra-episode success by 12.8 percentage points over the strongest baseline and achieves 78% subtask success on compositional generalization tasks, while using up to 4.5x less memory and running 2.5x faster than the keyframe baseline at long horizons.

cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02659

Parametric CAD Bench v3: Evaluating Agentic Parametric CAD Authoring from Text and Engineering Drawings

Mei Chen, Yu Xia, Haipeng Mao

原文摘要

Producing an executable CAD script is not sufficient to establish that an agent has satisfied an engineering design request. Evaluation must also consider the resulting geometry, specified dimensions, and native CAD modeling history. We present Parametric CAD Bench v3, a 100-task benchmark comprising 30 text-to-CAD creation tasks, 30 text-to-CAD create-and-edit tasks, and 40 tasks that reconstruct mechanical parts from engineering drawings. Agents operate in a pinned FreeCAD environment and submit native CAD artifacts to a separate programmatic verifier. The primary metric is mean continuous task reward, combining geometry similarity and specification consistency rather than binary task success. A September 24, 2026 snapshot covers 13 model-agent systems; Claude Code with Claude Opus 5.5 achieves the highest overall point estimate, 61.03%, and an image-to-CAD mean reward of 84.66%. These measurements describe complete systems, not isolated model capabilities. We release versioned tasks, evaluation configurations, and inspectable result artifacts.

cs.AI · cs.CV · cs.GR · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v5arXiv:2606.09570

UXBench: Benchmarking User Experience in AI Assistants

Mengze Hong, Xia Zeng, Zeyang Lei, Sheng Wang, Chen Jason Zhang, Di Jiang, Taiming Fu, Jinfeng Huang 等 16 位作者

原文摘要

As AI assistants serve millions of users daily, evaluating user experience (UX) beyond general model capability has become increasingly important. We present \textbf{UXBench}, the first user-centric benchmark grounded in real user feedback signals for evaluating preference alignment and dialogue generation. The benchmark consists of three interconnected tasks, UX Judge, UX Eval, and UX Recovery, with 7,400 test instances extracted from over 70K interaction logs of a mainstream Chinese AI assistant. The dataset closely reflects real user distributions, covering 8 scenarios, 83 domains, and diverse failure patterns that pose severe challenges. Extensive experiments on 26 frontier language models provide novel insights into how well models perceive user experience and how improvements in model capability contribute to better dialogue engagement. Through analyses of model behavior and performance gaps, we document six important findings, demonstrating that user feedback prediction is a learnable capability and revealing different aspects that influence user experience. UXBench establishes a new evaluation landscape and calls for greater attention to tailored UX optimization, contributing toward a user-centric scaling law for the development of successful AI assistants. The full project is released at https://github.com/mengze-hong/UXBench.

cs.CL · cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2510.04257

AgentTypo: Adaptive Typographic Prompt Injection Attacks against Black-box Multimodal Agents

Yanjie Li, Yiming Cao, Dong Wang, Bin Xiao

原文摘要

Multimodal agents built on large vision-language models are increasingly deployed in open-world settings but remain highly vulnerable to prompt injection. Typographic text-in-image attacks such as FigStep have been evaluated mainly on standalone vision-language models. For agents, user goals, webpage context, and action history make the effect of embedded instructions uncertain, and robustness to behavior-guided, adaptive visual injections remains insufficiently studied. We introduce AgentTypo, a black-box red-teaming framework that launches adaptive typographic prompt injection by embedding optimized text into webpage images. We formulate an attention-guided automatic typographic prompt-injection objective that favors the attacker's expected answer over the clean-page answer, together with task-conditioned visual attention from small question-answering models. A novel black-box optimization method searches text position, contrast, and color. To further strengthen the attack, we develop a multi-LLM system that iteratively refines injection prompts using evaluation feedback and retrieves successful past examples for strategy learning. Experiments on the VWA-Adv and Mind2Web benchmarks show that AgentTypo significantly outperforms state-of-the-art image-based prompt injection attacks. Across seven closed- and open-source model backends, AgentTypo-pro achieves 45%-83% ASR, outperforming the CLIP-based AgentAttack by 29-47% and previous typographic attacks by 25%-37% on VWA-Adv.

cs.CR · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02658

OctreeSplat: Hierarchical Octree-Guided Feed-Forward Gaussian Splatting

Yibin Zhao, Yihan Pan, Wenli Yang, Jun Nan, Jianjun Yi

原文摘要

3D Gaussian Splatting (3DGS) has shown considerable potential for novel view synthesis (NVS) and 3D reconstruction. Feed-forward methods avoid costly per-scene optimization, but fixed image or spatial resolutions limit their ability to allocate primitives according to local complexity. We present OctreeSplat, a hierarchical octree-guided feed-forward Gaussian splatting. OctreeSplat extracts multi-view features and geometric priors from a frozen visual foundation model (VFM), constructs a hierarchical octree, and predicts densification scores. The resulting representation allocates Gaussian primitives at spatially varying densities according to local complexity. Stochastic solids volume rendering improves depth rendering, strengthens geometric supervision, and supports surface reconstruction. When camera parameters are available, an optional test-time optimization (TTO) module improves agreement between predicted and ground-truth parameters. Training updates only 76 million additional parameters and requires 35 GPU-hours. Experiments on Replica, TUM, Tanks and Temples, and DTU demonstrate competitive RGB view synthesis, depth synthesis, and surface reconstruction. OctreeSplat reduces the number of Gaussian primitives while maintaining high reconstruction quality.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2604.19144

ReflectMT: Internalizing Reflection for Efficient and High-Quality Machine Translation

Kunquan Li, Yingxue Zhang, Zhibin Lan, Fandong Meng, Jinsong Su

原文摘要

Recent years have witnessed growing interest in applying Large Reasoning Models (LRMs) to Machine Translation (MT). While most approaches adopt a 'pre-thinking' paradigm and benefit from explicit reasoning trajectories, they suffer from substantial inference cost and latency. To address these limitations, we propose ReflectMT, a two-stage reflection internalization framework for machine translation that employs a 'post-thinking' paradigm. Our approach develops the model's 'translate-reflect-refine' capability through reinforcement learning. In the first stage, we cultivate the model's capacity for high-quality reflection and refinement, thereby enhancing its semantic comprehension and task-specific knowledge. In the second stage, we train the model to internalize the knowledge acquired during reflection. On WMT24, ReflectMT achieves a GPT-4o-based GRF score of 86.16, outperforming DeepSeek-R1 (84.03) by 2.13 points, while using approximately 92% fewer output tokens. Independent automatic evaluation, a blinded human evaluation, and preliminary experiments on three additional translation directions provide further evidence for the framework's effectiveness.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02657

Constructing an Artworld Influencers Network by Mining Social Media

Amalia Foka

原文摘要

The author shows a methodology for constructing a network map of the artworld. The tweets posted by a manually comprised set of exemplar artworld actors and Natural Language Processing (NLP) methods are used to identify significant artworld actors, including those that do not own a Twitter account. Identified artworld actors form communities detected with the Louvain method, and their authority is evaluated with the HITS centrality method. The resulting artworld network exhibits the ability, starting with a small set of exemplar actors, to identify new artworld actors and provides a wealth of data that would be very hard to collect otherwise.

cs.SI · cs.CL · cs.CY

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02654

VerdictAD: Few-Shot Anomaly Detection via Decision-Guided Refinement

Shangquan Sun, Wenjun Luo, Yuanzhi Wang, Ruiqi Li, Ruoyu Chen, Jiachen Liu, Shijian Lu

原文摘要

Detecting defects in real-world industrial inspection often requires learning from only a few normal images per category. Recent few-shot anomaly detection methods have achieved strong performance by measuring feature differences between query images and normal references. However, visual differences do not always indicate genuine defects, as normal appearance variations can also produce high anomaly scores. Conversely, structural or logical anomalies may not produce strong local feature differences, making them difficult to detect. To address this limitation, we introduce VerdictAD, a training-free framework that refines existing anomaly detectors through visual decisions. Rather than replacing existing detectors, VerdictAD adopts a typed decision model to assess the query image and its local regions against a few normal references. By directly producing structured decision probabilities without text generation, the model provides complementary evidence for refining anomaly responses at different spatial scales. Building on this capability, we introduce two modules: Global Verdict Refinement (GVR) uses image-level decisions to refine the overall anomaly score, while Local Verdict Refinement (LVR) aggregates regional decisions to refine the anomaly map. Combined with a simple DINOv2-based detector, VerdictAD achieves competitive performance in one-shot and few-shot anomaly detection. More importantly, as a plug-and-play framework, it consistently improves existing anomaly detectors across MVTec AD, VisA, and MVTec LOCO AD, achieving strong results across one-shot and few-shot settings. Our code and experimental logs are publicly available at https://github.com/sgstq0811/VerdictAD-Jev-AD.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02651

Probe-Guided Gradient Balancing for Multimodal Learning

Vu Vo, Haytham M. Fayek, Thuy Nguyen

原文摘要

Jointly trained multimodal networks frequently underutilize weaker modalities, as faster-learning modalities dominate the shared fusion objective and suppress gradients reaching slower ones. Most existing approaches throttle dominant modalities, whereas directly boosting weaker modalities often fails because monitoring signals derived from the joint loss remain dominated by stronger modalities. We propose Probe-Guided Gradient Boosting (PGGB), a gradient modulation approach that boosts weaker modalities while preserving the strength of dominant ones. PGGB attaches lightweight linear classifiers to stop-gradient features from each modality to estimate representation quality independently. The gap between these scores, termed the utilization gap, provides an imbalance signal used to adaptively scale weaker-modality gradients through bounded and smoothed modulation. PGGB is composable with throttling methods, as it leaves the loss function and fusion architecture unchanged. Experiments across eight benchmarks spanning four domains and 2-4 modalities demonstrate that PGGB outperforms state-of-the-art methods on highly imbalanced multimodal datasets while remaining competitive under low or negligible imbalance. We further establish bounded scaling, self-attenuation, and a descent bound for standard SGD under smoothness and bounded-variance assumptions. Code is available at https://github.com/VuVoCodes/PGGB.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.30365

Data-Void Exploits in Web-Searching LLMs

Michele Guerra, Wuyeh Jobe, Sanjana Nambiar, Christina Pöpper, Azza Abouzied

原文摘要

Web-searching Large Language Models (LLMs) have changed how users seek online information: instead of sifting through ranked lists of links, users read synthesized answers. This shift creates a new LLM vulnerability: a data-void exploit. Disinformers publish their narratives in search spaces where relevant and reliable content is scarce, i.e. data voids, then rely on the LLM's web-search to retrieve and then use, cite, or list those pages in its responses to data-void seeking prompts. Unlike classical search, where users see ranking cues and competing results, LLMs can obscure when an answer relies on sparse, unverified, or adversarial sources. We study this threat by varying the user's prompt tone - skeptical, excited, or neutral - and LLM choice - ChatGPT, Gemini, or Sonar Pro - as well as three attacker controls. The attacker can use one of five known strategies to construct and fill a data void with their narrative, host their content at varying levels of visibility from well-established, indexed domains to less established ones, and include generative-optimization engine features thought to boost citing by web-searching LLMs. We find that data-void exploits can succeed with only a single attacker controlled page. Across all conditions, attacker-content appears in LLM responses 31% of the time. User controls offer limited protection: while skeptical prompts slightly reduce success and slightly increase flagging or warnings in LLM responses, we find no evidence of large effects of prompt tone or model choice on success rates, although smaller effects cannot be ruled out. Overall, our results establish data voids as a practical inference-time attack surface for manipulating web-searching LLMs through strategically placed web content and highlight the urgent need for defenses against such exploits.

cs.CR · cs.HC · cs.IR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02649

Computer Vision Applications for Art History: Reflections and paradigms for future research

Amalia Foka

原文摘要

One of the contributing factors to the continuing debate among art historians over the use of computational methods in art history research is that they do not consider the core of today's art history research questions. The lack of close collaboration between the two involved research communities makes the definition of contemporary art-historical methods as well-defined computer vision problems extremely difficult. For that purpose, it is devised as a methodology to study articles in art history journals from a computer science perspective. The objective is to identify which image features art historians utilise within their research and describe them in immediate and meaningful terms to the computer vision research community. Finally, some paradigms that could serve as a new starting point for exploring how computer vision applications for art history can address the core of today's art history research are given.

cs.CV · cs.CY · cs.DL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02648

ReinSeg: Retrieval-Guided Bounded Multimodal Refinement for 3-D Road Anomaly Segmentation

Haifa Zhang, Yijing Wang, Zhiqiang Zuo

原文摘要

3-D road anomaly segmentation is essential for risk-aware perception in autonomous driving. Its central challenge is to rank sparsely observed hazards ahead of abundant normal road structures while preserving point-wise metric localization. LiDAR provides accurate 3-D point support, whereas surround-view imagery offers complementary semantic context for separating unexpected hazards from challenging normal structures. We propose ReinSeg, a retrieval-guided bounded multimodal refinement framework for point-wise road anomaly segmentation. Occupied LiDAR bird's-eye-view cells retrieve scene-specific context from a shared memory of eight surround-view images without supplying calibration variables to the retrieval module. A relative-norm-bounded residual returns this context to the LiDAR point features, so the frozen LIDO anomaly-scoring interface is preserved, and the refinement is optimized with in-distribution semantic labels only. On the Spotting the Unexpected (STU) benchmark, ReinSeg improves validation average precision from 27.54\% to 35.67\% and reduces the false-positive rate at 95\% anomaly recall from 34.85\% to 10.61\%, shrinking the high-recall screening set from 67.54 to 20.55 million points. Relative to REL, AP is 24.99 percentage points higher and FPR95 is 0.99 points lower. Matched interfaces and component isolation show that retrieved scene imagery and the relative-norm bound are both essential, and ReinSeg stays above LiDAR-only performance under missing, corrupted, or misaligned camera inputs while running at 21.8 frames per second. These results establish bounded retrieval of scene-specific visual evidence as an effective route to reliable 3-D road-hazard perception.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.05197

From Scientific Observations to Mechanisms: Benchmarking Hypothesis Generation by AI Scientists

Xiaxun Xie, Qingqing Long, Meng Xiao, Wei Ju, Yuanchun Zhou, Hengshu Zhu, Xuezhi Wang

原文摘要

Data-driven mechanistic hypotheses are essential to scientific discovery because they explain how underlying processes produce observed phenomena. AI agents and AI scientists increasingly support scientific data analysis. However, their ability to turn empirical findings into mechanistic hypotheses remains insufficiently examined. To address this gap, we introduce MechHypoBench, the first benchmark for evaluating whether AI agents and AI scientists can generate such hypotheses from empirical data. It combines paper-derived mechanisms from 14 scientific fields with real-world datasets containing 17.98 million records. The construction retains the observational complexity of empirical data while providing a specified underlying mechanism. Agents analyze the observations and propose open-form hypotheses. We develop an evaluation framework that assesses open-form mechanistic hypotheses through their consequences under withheld conditions. Experiments with general agents and AI scientists reveal a substantial gap between generated hypotheses and the underlying mechanisms.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.24202

When Does Multi-Agent RL Improve LLM Workflows? Workflow, Scale, and Policy-Sharing Tradeoffs

Yifan Zeng, Yiran Wu, Yaolun Zhang, Wentian Zhao, Kun Wan, Qingyun Wu, Huazheng Wang

原文摘要

Outcome-reward RL can train every role of a multi-agent LLM workflow. Its benefit is hard to predict because the reward assigns no credit to individual roles and a shared policy trades pooled data against specialization. We ask when this training improves a workflow over its base model and single-agent RL, and what each role learns. We train three workflows (Eval-Opt, Voting, Orch-Workers) with GRPO on math and code at three Qwen3 scales, with one policy for all roles (Shared-Policy, SP) or one per role type (Isolated-Policy, IP). Every workflow gains by adapting to its structure; this does not make a role better than training it alone. Single-agent RL in the same structure matches or beats trained Voting and Orch-Workers, and within the grid only Eval-Opt under IP exceeds it. Eval-Opt's evaluator learns when to retry and its generator learns to complete the retried attempts, a joint strategy that a single policy trained alone does not acquire. Under SP, per-output token averaging sets each role's gradient share, a hidden hyperparameter absent from single-agent RL, and the solving roles learn slowly; reweighting it gives SP the speed and peak of IP, and its decline. Under IP, a parallel role, whose instances share one adapter, takes one optimizer step per instance per training step and can drift after the peak as its entropy rises and accuracy declines. Merging those steps delays the decline. Our code is at: https://github.com/XHMY/marl-llm-workflows

cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02644

Coati: Host-Grounded Execution for Installable Mobile GUI Agents

Yudou Ying, Pu Shi, Hanlei Shi, Yuhua Zheng

原文摘要

Mobile GUI agents can carry out tasks across apps, yet most run on a separate computer that drives a phone over the Android Debug Bridge (ADB), out of reach of ordinary users and of the apps through which most people reach Web services. We present Coati, a source-available agent that users install as an Android app. Its runtime on the phone, the host, records what a model alone cannot see: the effect of each action at its target, result screens across apps, and the progress of repeated steps. VerifyLoop checks action outcomes and completion claims against host evidence and reports a completion it cannot audit as unverified. AppMemory builds per-app procedural memory from observations that the interface supports and reuses it on later visits. A routed fast executor lets a lightweight selection model take simple, confident steps and lets the host repeat loops of simple steps that the large model plans once. On all 116 AndroidWorld tasks, with Gemini 3.8 Flash as every agent's backbone, Coati succeeds on 86.5% of the tasks over three trials, ahead of the strongest open-source agent Artemis (77.6% with the same accessibility-only access, 76.5% even when allowed ADB) and far ahead of FinalRun and Midscene (about 59%), and VerifyLoop cuts false completion claims by about a third at comparable success. In everyday use, the fast executor cuts the median time of phone tasks by 39% at comparable success and runs three of four loop tasks 2.8-5.4x faster with 6-9x fewer model tokens, while AppMemory shortens a repeated deep route in an unfamiliar app from 18.5 steps to 6-7. On an Android 16 phone, routed execution completes all 27 benchmark runs with 76% fewer Reasoning calls on the critical path. Host evidence, not model self-assessment, makes an installable agent reliable and fast.

cs.MA

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02643

SearchInject: Retrieval Exposure Control for Indirect Prompt Injection in Language-Model Search

Kainan Zhou, Gangzhen Qian, Janet Sung, Zhi Jing, Rong Bao, Qianru Lao

原文摘要

An indirect prompt injection can affect a search-augmented large language model only after its carrier page reaches model context. SearchInject controls this admission decision with a joint score combining retrieval relevance, search-abuse cues, instruction likelihood, and two relevance interactions. A quarantine threshold is selected under an empirical 0.5% clean-candidate false-quarantine constraint on a calibration split. Evaluation uses 300 held-out SciFact queries and held-out payload categories within four construction families already represented during development. Across 1,200 query-family outcomes per retriever, top-10 target exposure falls from 0.9408 to 0.1583 for BM25, from 0.9158 to 0.1508 for a normalized hybrid, and from 0.8450 to 0.1375 for latent semantic analysis. These are relative reductions of 83.2-83.7%; the largest absolute change in clean normalized discounted cumulative gain at rank 10 is 0.0018. The frozen threshold yields clean test false-quarantine rates of 0.59-0.65%, above the empirical calibration constraint. A query-level analytical reanalysis accounts for the repeated construction outcomes under an independent-query assumption. A separate 104-case generation follow-up records context removal but no confirmed undefended hijacks under a post hoc endpoint, so it cannot estimate a reduction in attack success. The evidence supports retrieval exposure control within the evaluated construction families; the benefit of the interaction terms at matched clean cost remains unresolved.

cs.AI · cs.CY · cs.IR · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02640

InterFuse: Interaction-Guided Adaptive Fusion of Heterogeneous Models for Payment Risk Ranking

Haoyu Jiang, Weiyu Han, Ze Zhao, Jie Sheng, Ruobing Zuo, Xiao Tan, Shuai Chen, Tianyi Zhang

原文摘要

Payment platforms use view-specific models trained on transaction attributes, behavioral histories, and relational structures to rank transactions under limited review budgets. However, existing fusion methods often focus on prediction scores or interactions among model representations, with less attention to how these interactions should guide aggregation using each model's original prediction and input coverage. To address this limitation, we propose InterFuse, a fusion framework that uses a shared fusion state from cross-model interaction to guide transaction-adaptive aggregation. InterFuse encodes each model's representation, prediction score, and input coverage as a model-specific token, using learned states for unavailable outputs. Availability-aware attention updates these tokens and produces a shared fusion state. This state then guides aggregation over available models using their updated representations, original predictions, and input coverage. We evaluate InterFuse on an industrial payment replay and three public financial benchmarks using the same upstream models across fusion methods. InterFuse achieves the highest top-ranked recall and AUC on all four datasets, improving Recall@1% by 1.11-8.09 percentage points over the strongest baselines. Ablations show the benefits of cross-model interaction and adaptive aggregation, while industrial analyses find larger interaction gains when model predictions disagree. Additional analyses examine robustness to missing outputs and the trade-off between ranking gains and fusion latency.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2607.15880

Dynamics-Aware Meta-Imitation for Generalization to Unseen Robotic Manipulation

Zhenduo Shang, Xiyao Liu, Bohan Li, Xudong Wang, Teng Ren, Lianqing Liu, Zhi Han

原文摘要

Imitation Learning aims to learn skills from extensive observations and demonstrations for robots, so it suffers from data scarcity and environment generalization. The existing methods predominantly focus on imitation from in-domain tasks and consequently struggle with generalization to unseen tasks. To bridge this generalization gap, we propose the \textbf{D}ynamics-\textbf{A}ware \textbf{M}eta-\textbf{I}mitation (DAMI) framework. By integrating meta-learning to construct a shared skill space, DAMI equips agents for rapid adaptation to novel tasks. We introduce the Visual-Motor Trajectory (VMT) module to capture complex spatio-temporal dynamics within the task latent space. Furthermore, we propose the Unpaired Unified Task (U2T) block to fuse unstructured multimodal observations. To coordinate these representations, we integrate a Task-Conditioned Feature Modulation (TCFM) mechanism customized for modulating low-level 3D features. By capturing intrinsic dynamics from a random complete reference demonstration, our framework learns the underlying task logic rather than memorizing static cues, ensuring effective generalization. Extensive experiments in both simulation and real-world settings demonstrate that our approach outperforms state-of-the-art baselines regarding direct inference on seen tasks and adaptation to unseen tasks via few-shot fine-tuning.

cs.RO · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02639

Where Is a Prediction Model Miscalibrated, and for Whom? Local Audits in Outcome-Learned Pockets

Bernardo Modenesi, Taoyi Chen, Sima Najafzadehkhoei, Lucia Wang, Ameya Diwan, Beatrice Knudsen

原文摘要

Overall and group-level averages, and ranking metrics such as the AUC, can hide sets of observations for which a prediction model systematically over- or under-predicts. We present an audit that asks where a model is miscalibrated (in which sets of similar observations it over- or under-predicts on average) and for whom (whether its average prediction error differs between demographic groups, such as by race or sex, within the same set). It groups observations into pockets using random-forest similarity, which weighs each feature by its importance for predicting the outcome, with a ridge-regression grouping alongside. Pockets are learned on one sample and tested on another, with simultaneous confidence intervals that allow for correlation within units such as hospitals; the audit needs only features, outcomes and fixed predictions. In 17 simulation scenarios with an error planted for one group, the combined audit located it better than the strongest alternative grouping in 15, with false-alarm rates near or below 5%; with a local error planted for everyone, it won in 12 of 15 scenarios at the smaller error size and lost when the error was unrelated to the outcome. In public synthetic data from the care-management algorithm studied by Obermeyer et al., it found no detectable overall error but under-prediction of illness in the sickest pockets and over-prediction elsewhere, and distinguished Black patients' concentration where the score under-predicted for everyone from a Black-White gap within pockets. For APACHE IVa, an exploratory analysis located its previously reported over-prediction of ICU mortality in a mostly ventilated pocket and a low-risk pocket, with none detected in a moderate-risk one. Two chest X-ray classifiers showed offsetting local errors behind no detectable overall error. Code: https://github.com/modenesi/pocket-audit

cs.LG · cs.AI · stat.ME · stat.ML

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2601.12428

ReWorld: Multi-Dimensional Reward Modeling for Embodied Video Generation

Baorui Peng, Wenyao Zhang, Liang Xu, Bohan Li, Hu Zhu, Zekun Qi, Hongsi Liu, Jiazhao Zhang 等 10 位作者

原文摘要

Current embodied video generation models primarily emphasize visual generative quality while overlooking physical fidelity, dynamic consistency, and task logic, especially for contact-rich manipulation tasks, which limits their applicability to downstream tasks. In this paper, we introduce ReWorld, a framework aimed to employ reinforcement learning to align the video generation with physical realism, task completion capability, embodiment plausibility and visual quality. Specifically, we first construct a large-scale (~235K) video preference dataset and employ it to train a hierarchical reward model designed to capture multi-dimensional reward consistent with human preferences. We further propose a practical alignment algorithm that post-trains flow-based video generators using this reward through a computationally efficient PPO-style algorithm. Comprehensive experiments and theoretical analysis demonstrate that ReWorld significantly improves the physical fidelity, logical coherence, embodiment and visual quality of generated rollouts, outperforming previous methods.

cs.RO · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02637

Variable names provide priors: semantic grounding enables data-efficient numerical prediction by large language models

Xin-Rui Ma, Shi-Zhi Chen, Hai-Yan Jiang, De-Cheng Feng, Michael Beer

原文摘要

Large language models (LLMs) have achieved strong performance across diverse tasks, from question answering to code generation. Recent studies have found that LLMs can perform numerical prediction on tabular datasets via in-context learning. However, it remains unclear under what conditions and how LLMs use variable names and in-context examples for numerical prediction. Here we show that semantic grounding, the match between variable names and the values they describe, allows five LLMs without task-specific training to match or outperform seven supervised machine learning (ML) models across 20 datasets from nine scientific domains, particularly when examples are scarce. Reasoning traces indicate that variable names allow LLMs to draw on domain-specific knowledge acquired during pretraining. Using controlled tasks based on physical formulas, we further show that LLMs integrate domain-specific knowledge with in-context examples in a way consistent with Bayesian updating. Specifically, when the relationship between the inputs and the target cannot be inferred from the examples alone, LLMs rely on domain-specific knowledge as the prior. Conversely, as contradicting examples accumulate, LLM predictions shift progressively towards these examples, which act as the likelihood. Our work reveals how LLMs use domain-specific knowledge in numerical prediction and positions them as data-efficient predictors for data-scarce scientific domains.

cs.AI · cs.CL · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02634

Evolving Explanations: Complexity-Adaptive Refinement of Faithful Explanations for Symbolic Regression

Elisabeth Mayrhuber, Ting Hu, Stephan Winkler

原文摘要

Natural-language explanations generated by large language models (LLMs) for machine learning models often seem fluent but are not necessarily faithful to the underlying model. We formalize this mismatch as a faithfulness gap and propose a complexity-adaptive refinement framework for symbolic regression that iteratively generates and validates explanations until numerical agreement with the target expression is achieved. The approach follows a two-stage pipeline: an explainer LLM produces textual descriptions at increasing levels of detail, while an independent validator LLM extracts a symbolic expression that is compared against the original model. This process is framed as a discrete search over explanation space, where complexity is minimized subject to a faithfulness constraint. We evaluate the method on real-world and synthetic datasets, with a total of 250 runs across varying model complexities. The explanation refinement loop achieves an overall pass rate of 84.0%, with most explanations converging at low complexity levels and only a small fraction requiring detailed descriptions. Our results demonstrate that faithful explanations can be obtained with minimal necessary complexity by explicitly closing the explanation loop through symbolic validation.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02633

Measuring AI Agents as Teammates with Collaborative Intelligence

Sean Wu, Aneesh Pappu, Jiaru Zou, Yubin Kim, Yuncong Mao, Hongjian Zhou, Junchi Yu, Jieneng Chen 等 9 位作者

原文摘要

Progress in AI is often measured by what a single model can do on its own. Yet teams of AI agents are starting to make breakthroughs and discoveries in science, medicine, mathematics and software engineering. As AI begins to work in teams of agents and people, the ability to collaborate becomes a core part of their intelligence. Research on human teams has measured this ability for decades. Inspired by this, we introduce the Collaborative Intelligence Benchmark, which measures how well an AI agent works as a teammate. It contains 240 tasks across 8 dimensions of teamwork, from refusing an unsafe shortcut to checking the work that a weaker model leaves behind. Each task comes from a real record or an existing benchmark, from aviation accident reports and clinical cases to coding tasks and negotiations, and has a programmatically verifiable outcome. To isolate teamwork from individual capability, models are scored only on tasks that they can solve by themselves when they have full information. We evaluate 18 closed- and open-weight models and find that every model, across sizes and families, has clear weaknesses as a teammate, with the best model scoring only 82.2 out of 100. Frontier models almost always refuse unsafe shortcuts and report misconduct, yet they cave under interpersonal pressure, reaching the best overall deal in only 14% of conflicts, and their pass rate drops from 97% to 81% when they take over a task in which a weaker model has cheated. Crucially, general capability does not guarantee good teamwork, as models with nearly the same capability score differ by up to 11 points as teammates. The benchmark includes 160 public tasks with code and data, along with 80 held-out tasks for evaluation. As AI agents increasingly work alongside other agents and people, the ability to collaborate effectively is becoming a crucial capability to measure and develop.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02632

Susceptibilities for Neural Networks Learning from Physical Data

Rohan Hitchcock, Gary W. Delaney, Jonathan H. Manton, Richard Scalzo, Jingge Zhu

原文摘要

Neural networks that are trained on data arising from a physical system must somehow learn regularities induced by the underlying physical laws. In this setting, concepts from statistical physics provide powerful tools for analysing how physical laws influence the learning process. In this paper we explain how susceptibilities --- which measure the response of a neural network to perturbations of a parameter of the data distribution --- can be applied when learning from physical data. We show that such susceptibilities can identify the conditions under which specific input features are most informative for learning, guiding data selection, a prediction we validate experimentally. For distributions lacking a natural parameter, we propose introducing one via a physically meaningful scalar function of the physical state, and demonstrate that susceptibilities track the emergence of specific capabilities (such as collision detection) during training.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02631

UniCross3D: Unified Cross-View and Cross-Domain Diffusion for Consistent Single-Image 3D Generation

U-Chae Jun, Jaeeun Ko, Jiwoo Kang

原文摘要

Reconstructing detailed geometry and realistic appearance from a single RGB image is essential yet fundamentally challenging due to inherent ambiguities such as occlusion, lighting variations, and texture-geometry entanglement. While recent diffusion-based generative models have significantly improved novel view synthesis, existing approaches suffer from two critical limitations: lack of cross-view geometric consistency and insufficient cross-domain semantic alignment. To address these issues, we introduce UniCross3D, a unified cross-view and cross-domain diffusion framework designed explicitly for consistent and physically coherent 3D generation. UniCross3D features two novel contributions: (1) a cross-view latent regularization that enforces cross-view geometric consistency across synthesized viewpoints by penalizing latent variance, and (2) a cross-domain mutual information objective grounded in the physics of image formation, explicitly aligning synthesized color and normal maps. Extensive experiments demonstrate that UniCross3D achieves significantly improved view consistency and semantic alignment over state-of-the-art methods and yields higher-fidelity reconstructions, particularly under challenging textures and ambiguous viewpoints.

cs.CV · cs.GR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02630

ICJ-Graph: A Citation Graph and a Layered Document Corpus of the International Court of Justice

Jirameth Wannasiwaporn, Pawitsapak Akarajaradwong, Chompakorn Chaksangchaichot

原文摘要

We present ICJ-Graph, a machine-readable resource connecting documents of the International Court of Justice (ICJ) to cited cases, legal provisions and United Nations resolutions. Court decisions, written pleadings and oral records form distinct document layers, preserving the distinction between the Court's reasoning and the parties' submissions. The collection contains 6,396 ICJ documents and 505 documents from other courts and legal bodies, together with 12,375 provisions and 20,291 UN resolutions. The data files contain text, metadata and embeddings. The accompanying pipeline lets users load the graph and generate embeddings for use in their own retrieval systems. To illustrate the resource's use, we evaluate five models on citation prediction for 199 decisions using documents available before each decision. Tools designed for citation research improve citation-prediction F1 over answering without tools for all five models across case, provision and resolution citations.

cs.CL

查看论文
arXiv ↗PDF ↗