让下一篇重要论文，不再错过。

版本更新 · v2arXiv:2610.21180

更新于 2026-11-05

Evo-DexBench: Benchmarking Generalization across Related Tasks for Dexterous Manipulation

Bing Cheng, Yiran Mao, Zewei Ye, Yifan Jin, Xinyi Chen, Ruifeng Tu, Shuhuai Wang, Tao Lin 等 13 位作者

原文摘要

General-purpose dexterous manipulation requires policies to transfer knowledge from related tasks to new tasks with only a few target demonstrations. Existing benchmarks often provide too many target demonstrations and leave the differences between source and target tasks uncontrolled. As a result, they cannot clearly distinguish knowledge transfer from task memorization. Our key insight is the Core-Gen principle. Core tasks are related tasks that share a functional interaction and provide source knowledge, while a Gen task is a novel target within the same functional family that has only a few demonstrations. This setup tests whether a policy can reuse functional interaction knowledge. Based on this principle, we propose Evo-DexBench, which controls both source-task relatedness and target-task exposure. It comprises 30 tasks in six functional families, three everyday scenes per task, and 1,248 multimodal demonstrations collected via VR teleoperation with simulated Shadow Hands equipped with 123 taxels. Experiments show that, among six representative policies, even the best-performing method achieves an average success rate of only 24.4%, highlighting the challenges existing methods face in transferring functional interaction knowledge. Further analysis shows that, across the two tasks, joint Core + Gen training improves performance by 12% and 38%, respectively, over training on Core followed by fine-tuning on Gen, indicating that joint training makes better use of Core knowledge. With matched Core data, 8 Gen demonstrations achieve performance comparable to training on 16--32 Gen demonstrations alone. On all three tested Gen targets, functionally matched Core data yields greater gains in success rate than mismatched data. We release the Evo-DexBench dataset and benchmark to support the evaluation of dexterous manipulation policies' generalization across related tasks.

cs.RO

新论文 · v1arXiv:2611.06682

更新于 2026-11-04

SecureGARD: Dynamic Ontology-Guided Governance for Secure Tool-Using LLM Agents

Guoqin Xiong, Shengtao Wen, Xiang Chen, Xinyi Zeng, Zhaoxin Fan, Wei Ye, Xiaofeng Cao, Sheng-Jun Huang

原文摘要

Tool-using large language model (LLM) agents translate model outputs into consequential actions, turning indirect prompt injection into an execution-security problem. Existing defenses constrain tool calls, trajectories, or information flows, but often leave action semantics, state dependencies, and evidence relations implicit, limiting generalization across heterogeneous tools and complicating safe updates. We present \ours{}, a dynamic ontology-guided governance framework. \ours{} grounds heterogeneous calls in a shared semantic representation and combines an executable ontology with episode-level task state to derive deterministic ready, waiting, and forbidden decisions. A waiting decision identifies an unmet but recoverable obligation, enabling bounded replanning without expanding the agent's authorization boundary. \ours{} evolves its ontology offline from verified training executions. An offline LLM proposes instance-independent action--state axioms, which are accepted only if they address ontology errors exposed by training failures, preserve logical consistency and protected entailments, and pass held-out validation before the ontology is frozen. We evaluate \ours{} on AgentDojo, ASB, and AgentDyn under benign and indirect prompt-injection settings. \ours{} achieves a 0\% attack success rate (ASR) across all three benchmarks while improving task utility under attack compared with existing baselines. These results show that explicit semantic governance can strengthen execution security without indiscriminate blocking.

cs.LG

新论文 · v1arXiv:2611.06160

更新于 2026-11-04

DSRS: Teaching MLLMs to Perceive, Reason, and Segment in Complex Driving Scenes

Sicen Guo, Zhengfei Song, Kuanhsun Chen, Xu Huang, Ping Zhong, Wei Ye, Shaoyi Du, Rui Fan

原文摘要

Existing reasoning segmentation frameworks typically employ multimodal large language models (MLLMs) with broad world knowledge to extract semantics in a human-like manner and generate corresponding segmentation masks. Nevertheless, these frameworks lack the necessary scene-specific awareness. This paper introduces driving scene reasoning segmentation (DSRS), a scene-specific task that requires models to interpret implicit instructions, infer spatial relations, detect potential risks, handle challenging conditions, and comprehend underlying traffic rules, ultimately generating accurate segmentation masks. Meanwhile, we establish a benchmark dataset to facilitate future research on the DSRS task. To pioneer this task, we propose DSRS-Net, a novel architecture with advanced reasoning segmentation capability for automated driving. Specifically, DSRS-Net seamlessly infuses geometric features for deeper reasoning segmentation, without compromising the original multimodal fusion mechanism. This strategy effectively mitigates the modality representation mismatch in MLLMs caused by RGB-only pretraining, while enhancing the integration of geometric, contextual, and linguistic features. Extensive experiments demonstrate the superior performance of DSRS-Net in both driving scene reasoning segmentation and other language-driven segmentation tasks across various public datasets. The source code and dataset will be released upon publication.

cs.CV

版本更新 · v3arXiv:2507.16271

更新于 2026-11-03

Beyond Isolated Dots: Benchmarking Structured Table Construction as Deep Knowledge Extraction

Tianyun Zhong, Guozhao Mo, Yanjiang Liu, Yihan Chen, Lingdi Kong, Xuanang Chen, Yaojie Lu, Hongyu Lin 等 12 位作者

原文摘要

Large language models (LLMs) are increasingly expected to extract and synthesize structured information from complex real-world documents, yet they typically produce unstructured, hard-to-verify paragraph-style outputs. We introduce AOE (Arranged and Organized Extraction), a bilingual benchmark that evaluates LLMs' ability to distill long-form, multi-document inputs (64k to 1.8M tokens) into structured tables. AOE comprises 12 tasks across four domains (academic, financial, legal, and medical), each grounded in genuine analytical needs, with schema guidance provided to isolate extraction quality from formatting variance. We evaluate 12 frontier LLMs (both open-weight and proprietary) under four generation settings of increasing autonomy: direct full-document input, RAG-based retrieval, offline document-browsing agent, and online web-search agent. Even the best-performing model achieves only 42.7% Cell F1, and retrieval-augmented approaches often degrade rather than improve performance, revealing that structured extraction from long contexts remains a fundamental challenge - an "Area of Effect" (AOE) that impacts all tested systems.

cs.CL

新论文 · v1arXiv:2611.04285

更新于 2026-11-03

Optimal Communication for Distributed Sparse Histogram Aggregation

Fangwei Ye, Yijun Fan

原文摘要

Histogram estimation is a fundamental primitive in distributed data analysis, with applications in counting queries and federated learning. In many settings, the universe of categories is enormous, yet each user's data involves only a small subset, yielding sparse local histograms. This raises a natural question: To what extent can local sparsity reduce the communication needed to estimate the global histogram? We answer this question by characterizing the communication complexity of distributed sparse histogram estimation under prescribed $\ell_1$ distortion and failure probability constraints. Our upper bound is achieved by a randomized covering scheme that optimizes how accurately each user represents its local histogram to reduce communication. By exploiting the distortion budget directly rather than imposing unbiasedness, the scheme achieves lower communication costs than standard unbiased sampling and hashing constructions. We establish the scheme's first-order optimality through a lower bound based on local support identification. The key step is to relate communication cost directly to the number of active coordinates revealed by each user's message, rather than to bound information about the aggregate via distance-based Fano's inequalities. This characterization further quantifies how the communication advantage of heterogeneous over homogeneous encoders varies across parameter regimes.

cs.IT

新论文 · v1arXiv:2611.03850

更新于 2026-11-03

RoboScope Kit: Unified Interfaces for Fast and Scalable Embodied-AI Evaluation

Deng-Ke Wu, Yifan Jin, Yida Yang, Lei Qian, Zhaoye Zhou, Hongru Zhang, Kai Ye, Shuyuan Hu 等 11 位作者

原文摘要

Simulation benchmarks are the standard way to compare robot policies, but each benchmark brings its own evaluator, observation layout, and action convention, and evaluation is slow: 500 RoboTwin Easy episodes with one simulator and one model on one GPU take 359.7 minutes. Profiling the official StarVLA evaluator shows that simulation occupies CPU cores while the GPU averages 14.3% utilization. RoboScope Kit wraps each model once in a Python class called a Policy, and each benchmark once in an adapter around the benchmark's own evaluator, so the benchmark's rules and scoring stay unchanged. Scheduling runs simulator workers and Policy replicas as separate pools, sizes both from measured probes, and shards work deterministically across GPU instances; runtime optimization renders only the observations an action-chunk Policy reads. Adapters cover six benchmark families. On LIBERO-Long with a StarVLA checkpoint on one 48 GiB RTX 4090, the concurrency search selects twelve simulator workers sharing three Policy replicas, which evaluate 500 episodes 4.45$\times$ faster than the official StarVLA evaluation with equivalent success rates, 2.78$\times$ faster than RLinf, and 1.52$\times$ faster than vla-eval at its tuned configuration. With CUDA MPS, which lets the replicas run their kernels concurrently, the search selects sixteen workers sharing four replicas, 6.16$\times$ faster than StarVLA, 3.85$\times$ faster than RLinf, and 2.10$\times$ faster than vla-eval. On RoboTwin Easy, four GPUs with chunk-boundary observation complete the 500 episodes in 14.4 minutes, 25.04$\times$ faster than the single-GPU baseline, with identical success outcomes and terminal steps.

cs.RO · cs.AI

版本更新 · v4arXiv:2604.11674

更新于 2026-10-29

AffordSim：面向可供性感知机器人操作的可扩展数据生成器与基准

AffordSim: A Scalable Data Generator and Benchmark for Affordance-Aware Robotic Manipulation

Mingyang Li, Haofan Xu, Haowen Sun, Xinzhe Chen, Sihua Ren, Liqi Huang, Xinyang Sui, Chenyang Miao 等 13 位作者

AI 速读基于摘要

AffordSim 将开放词汇 3D 可供性预测融入仿真轨迹生成，构建 50 任务基准，并验证生成数据可零样本迁移到真实 Franka FR3。

cs.RO · cs.AI

查看原文摘要

Many everyday robot manipulation skills are affordance-dependent, with success determined by whether the robot contacts the functional object region required by the subsequent action. Current simulation data generators obtain contacts from generic grasp estimators or per-object manual contact annotations, but generic estimators rank stable grasps without task semantics and often select contacts that are misaligned with the downstream action, while manual contact annotations must be rewritten for each new object and task. To solve these challenges, we introduce AffordSim, a scalable data generator and benchmark that integrates open-vocabulary 3D affordance prediction into simulation-based trajectory generation. Given a natural-language task description, AffordSim synthesizes a task-relevant scene, emits affordance queries, grounds them on object surfaces, samples region-conditioned grasps, and selects executable candidates with motion planning. It further randomizes object pose, texture, lighting, image noise, and cross-viewpoint backgrounds for sim-to-real transfer. We instantiate AffordSim as a 50-task benchmark across diverse manipulation skills, five robot embodiments, and 500+ rigid and articulated objects. AffordSim achieves 93% of the trajectory collection success rate of manual contact annotations on affordance-critical tasks and 89% on hard composite tasks. Vision-language-action policies trained on AffordSim data transfer zero-shot to a real Franka FR3, reaching 24% average success.

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-demo.73ACL Anthology · ACL 2026

来源日期 2026-07-01（仅到月份）

Ryze: Evidence-Enriched Data Synthesis from Biomedical Papers

Yeqi Huang, Yue Chen, Yanwei Ye University of Edinburgh, University of Edinburgh, Guanhao Su, Luo Mai University of Edinburgh

原文摘要

General-purpose VLMs remain unreliable for biomedical research because valid answers in scientific papers depend on evidence split across figures, tables, charts, captions, and referring text.Existing post-training pipelines are bottlenecked by costly expert annotation and by synthetic data that drops this evidence structure.We present Ryze, a fully automated system that converts raw biomedical papers into an evidence-enriched training set and a domain-specialized VLM.Ryze synthesizes QA pairs with complete supporting evidence (visual element, caption, extracted structure, and referring paragraphs), reduces layout and OCR errors via chart/table-aware extraction and LLM-based cleansing, and applies a two-stage post-training strategy combining supervised fine-tuning with reinforcement learning.Starting from Qwen3-VL-8B, Ryze produces BioVLM-8B at under $200, achieving 48.0% weighted accuracy on LAB-Bench—outperforming the base model by +12.6% and surpassing GPT-5.2 by +3.8%.We release Ryze as open source together with the trained BioVLM-8B model.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1045ACL Anthology · ACL 2026

来源日期 2026-07-01（仅到月份）

A Survey of Reinforcement Learning for Large Language Models under Data Scarcity: Challenges and Solutions

Zhiyin Yu, Yuchen Mou, Juncheng Yan, Junyu Luo, Chunchun Chen Tongji University, Xing Wei, Yunhui Liu Nanjing University, Hongru Sun University of Wollongong 等 20 位作者

原文摘要

Reinforcement learning (RL) has emerged as a powerful post-training paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, reinforcement learning for LLMs faces substantial data scarcity challenges, including the limited availability of high-quality external supervision and the constrained volume of model-generated experience. These limitations make data-efficient reinforcement learning a critical research direction. In this survey, we present the first systematic review of reinforcement learning for LLMs under data scarcity. We propose a bottom-up hierarchical framework built around three complementary perspectives: the data-centric perspective, the training-centric perspective, and the framework-centric perspective. We develop a taxonomy of existing methods, summarize representative approaches in each category, and analyze their strengths and limitations. Our taxonomy aims to provide a clear conceptual foundation for understanding the design space of data-efficient RL for LLMs and to guide researchers working in this emerging area. We hope this survey offers a comprehensive roadmap for future research and inspires new directions toward more efficient and scalable reinforcement learning post-training for LLMs.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1628ACL Anthology · ACL 2026

来源日期 2026-07-01（仅到月份）

LeLoRA: Learnable Low-Rank Adaptation of Large Language Models

Xiaoling Zhou, Mingjie Zhang, Zhemg Lee, Wei Ye Peking University, Shikun Zhang Peking University

原文摘要

Fine-tuning large language models (LLMs) is an effective approach to enhancing their performance on specialized downstream tasks. Among the various techniques, low-rank adaptation has garnered significant attention due to its ability to maintain the full performance of fine-tuning while enhancing computational efficiency. However, existing approaches often rely on manually specified and fixed hyperparameters to identify the trainable components within weight matrices, resulting in suboptimal performance and low parameter efficiency. This paper presents a novel Learnable Low-Rank Adaptation (LeLoRA) framework that utilizes dynamically learned fine-tuning strategies to facilitate the effective adaptation of LLMs. Our framework integrates an LLM with a policy network that automatically and adaptively generates matrix-specific adaptation strategies to identify the trainable components of each weight matrix, taking into account their unique characteristics, such as singular values and matrix norms. A reinforcement learning-based optimization algorithm is then employed to iteratively update the LLM and the policy network, ensuring that the generated strategies adapt in real time to the evolving states of the LLM. Extensive experiments have been conducted across various natural language processing and multimodal tasks. The results across ten different LLMs, ranging from 125M to 70B parameters, provide compelling evidence that LeLoRA consistently outperforms existing baselines in adapting LLMs. Moreover, analytical experiments provide valuable insights into the effectiveness of the generated strategies.

cs.CL · cs.CV

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.196ACL Anthology · ACL 2026

来源日期 2026-07-01（仅到月份）

Retrieval as Generation: A Unified Framework with Self-Triggered Information Planning

Bo Li Peking University, Mingda Wang Hebei University of Technology, Gexiang Fang, Shikun Zhang Peking University, Wei Ye Peking University

原文摘要

We revisit retrieval-augmented generation (RAG) by embedding retrieval control directly into generation. Instead of treating retrieval as an external intervention, we express retrieval decisions within token-level decoding, enabling end-to-end coordination without additional controllers or classifiers. Under the paradigm of Retrieval as Generation, we propose GRIP (Generation-guided Retrieval with Information Planning), a unified framework in which the model regulates retrieval behavior through control-token emission. Central to GRIP is Self-Triggered Information Planning, which allows the model to decide when to retrieve, how to reformulate queries, and when to terminate, all within a single autoregressive trajectory. This design tightly couples retrieval and reasoning and supports dynamic multi-step inference with on-the-fly evidence integration. To supervise these behaviors, we construct a structured training set covering answerable, partially answerable, and multi-hop queries, each aligned with specific token patterns. Experiments on five QA benchmarks show that GRIP surpasses strong RAG baselines and is competitive with GPT-4o while using substantially fewer parameters. Code and resources are provided in the supplementary materials.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.214ACL Anthology · ACL 2026

来源日期 2026-07-01（仅到月份）

Instruction Data Selection via Answer Divergence

Bo Li Peking University, Mingda Wang Hebei University of Technology, Shikun Zhang Peking University, Wei Ye Peking University

原文摘要

Instruction tuning relies on large instruction–response corpora whose quality and composition strongly affect downstream performance. We propose Answer Divergence-Guided Selection (ADG), which selects instruction data based on the geometric structure of multi-sample outputs. ADG draws several high-temperature generations per instruction, maps responses into an embedding space, and computes an output divergence score that jointly encodes dispersion magnitude and shape anisotropy. High scores correspond to instructions whose answers are both far apart and multi-modal, rather than clustered paraphrases along a single direction. Across two backbones and three public instruction pools, fine-tuning on only 10K ADG-selected examples consistently outperforms strong selectors on six benchmarks spanning reasoning, knowledge, and coding. Analyses further show that both dispersion magnitude and shape anisotropy are necessary, supporting answer divergence as a practical signal for instruction data selection. Code and appendix are included in the supplementary materials.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.501ACL Anthology · ACL 2026

来源日期 2026-07-01（仅到月份）

Learning from Contrasts: Synthesizing Reasoning Paths from Diverse Search Trajectories

Peiyang Liu, Zhirui Chen, Xi Wang Px Securities, Di Liang, Youru Li Beijing University of Technology, Zhi Cai Beijing University of Technology, Wei Ye Peking University

原文摘要

Monte Carlo Tree Search (MCTS) has been widely used for automated reasoning data exploration, but current supervision extraction methods remain inefficient. Standard approaches retain only the single highest-reward trajectory, discarding the comparative signals present in the many explored paths. Here we introduce Contrastive Reasoning Path Synthesis (CRPS), a framework that transforms supervision extraction from a filtering process into a synthesis procedure. CRPS uses a structured reflective process to analyze the differences between high- and low-quality search trajectories, extracting explicit information about strategic pivots and local failure modes. These insights guide the synthesis of reasoning chains that incorporate success patterns while avoiding identified pitfalls. We show empirically that models fine-tuned on just 60K CRPS-synthesized examples match or exceed the performance of baselines trained on 590K examples derived from standard rejection sampling, a 20\times reduction in dataset size. Furthermore, CRPS improves generalization on out-of-domain benchmarks, demonstrating that learning from the contrast between success and failure produces more transferable reasoning capabilities than learning from success alone.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表