跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+0已有论文修订
+91新增 AI 导读

arXiv 最近完成扫描:2026-11-22 16:55 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-21 20:23 — 2026-11-22 20:23(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:2026-11-22 16:55最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 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。

29,158 已收录论文24,181 AI 相关8,992 有修订112 HF 社区发现

浏览论文库

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 73 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-22 16:52(北京时间)
新论文 · v1arXiv:2611.26177

SpatialScout-V2: Can VLM Agents Actively Explore Real-World 3D Scenes and Localize Objects?

Zhaoxi Zhang, Hao Liang, Linxuan Wu, Yiming Xu, Yifu Yuan, Kun Liang, Chenming Tang, Meiyi Qiang 等 11 位作者

原文摘要

Spatial intelligence requires agents to acquire visual evidence and integrate partial observations into coherent three-dimensional representations. We introduce SpatialScout-V2, which extends SpatialScout's active mapping task from simulation to real residential scenes through three advances: real visual observations, preservation of original scene content, and interactive exploration of recorded views. Built from 100 homes in Realsee3D, the benchmark covers 3,171 target objects across 29 categories. Given a fixed target inventory, agents autonomously select recorded viewpoints and viewing directions, integrate calibrated RGB observations, and iteratively estimate and revise object centers within a fixed interaction budget. We evaluate the localization accuracy and completeness of object-center maps using mean IoU, IoU-threshold hit rates, and target coverage. Evaluating seven frontier vision-language models (VLMs), we find that six achieve an all-target mean IoU below 20\%. Even the best-performing model, Claude Opus 5.5, reaches only 22.52\%, with just 18.36\% of targets exceeding an IoU of 0.5. Further analysis shows that broad submission coverage can coexist with limited localization accuracy, while target omissions further reduce overall scores. These results highlight substantial challenges in constructing accurate and complete object-center maps through active exploration of real scenes.

cs.CV

新论文 · v1arXiv:2611.26075

Understanding MoE Decisions Beyond Gradients

Zhen Xu, Zitao Zhang, Kevin Zhao, Xuefeng Liu

原文摘要

Understanding a language model's decision requires knowing how its internal computations affect its preference between answers. Gradient-based methods estimate this influence by predicting the effect of changing those computations. In mixture-of-experts (MoE) models, replacing one expert can change which experts later layers select. Those switches can alter the answer scores in ways that derivatives do not capture. We introduce a predictor that uses the model's own expert outputs and selection rules to estimate these extra effects, then adds them to a second-order prediction. Across questions from three benchmarks, our method reduces the error in predicting answer-score changes by 60\% on Qwen3.6 and 47\% on Gemma4, relative to a multistep baseline using gradients and second derivatives.

cs.CL · cs.LG

新论文 · v1arXiv:2611.25435

VisualJevBench: A Benchmark and Black-Box Attacks for Visual Typed Decision Models

Li Zeng, Yeqi Zeng, Hengkai Liang, Shiguo Wang, Meng Xie, Zeyu Ye, Jinheng Zhang, Xianlong Wang 等 10 位作者

原文摘要

Typed decision models (TDMs) return structured decisions instead of generating free-text responses, enabling fast and cheap decisions in real-world applications. Recent support for image inputs extends Jev, a representative TDM, to visual decision-making. We refer to this setting as Visual Jev. However, red teaming of Visual Jev leaves two gaps: a systematic evaluation of direct image manipulations is lacking, and universal visual attacks remain underexplored. To address these gaps, we introduce VisualJevBench, a benchmark for evaluating the adversarial robustness of Visual Jev. It comprises 800 images and 3,200 questions across eight scenarios and four visual attack surfaces: typographic manipulation, visual evidence spoofing, target selection interference, and geometric transformation. We also introduce Adaptive Black-Box Parameter Search (ABPS), a lightweight framework that uses typed-output feedback to find effective image manipulations. ABPS supports both image-specific attacks and universal attacks that transfer to unseen images and tasks in new scenarios without further optimization. Extensive experiments on jev-1.14.0 show that image-specific attacks achieve an attack success rate (ASR) of up to 69.6%, while universal attacks achieve an average ASR of 33.15% on unseen images. Geometric transformations achieve an average ASR of 40.09% in cross-scenario transfer.

cs.CR

新论文 · v1arXiv:2611.24643

AggCLIP: Aggregating CLIP for Zero-Shot Multimodal Anomaly Detection

Zhi Li, Hanyu Wang, Fei Shen, Zhengtao Zhang

原文摘要

Zero-Shot Multimodal Anomaly Detection (ZSMAD) aims to identify both appearance and geometric defects in novel categories without relying on target-specific training data, making it especially valuable in real-world scenarios constrained by data scarcity, privacy concerns, or high annotation costs. While recent approaches successfully adapt vision-language models like CLIP to achieve this open-vocabulary generalization, they still face significant challenges: (i) severe geometric information loss during 3D-to-2D point cloud rendering; (ii) spatially noisy 2D feature maps produced by raw CLIP patch embeddings, compounded by shallow late-stage modality fusion; and (iii) severe cross-modal semantic misalignment due to decoupled processing streams. To address these limitations, we propose AggCLIP, a novel framework that adapts frozen CLIP representations for ZSMAD by explicitly incorporating local geometric structure and visual spatial context. Specifically, to accurately localize fine-grained defects, we design a Geometry-Aware Spatial Aggregation (GSA) module that refines back-projected 3D point features via density-adaptive neighborhood weighting, alongside a Context-Aware Spatial Aggregation (CSA) module that smooths multi-scale 2D patch embeddings using a lightweight trainable adapter. Furthermore, to prevent semantic drift between modalities, we introduce a Dual-Stream Prompt Contrastive (DSPC) loss, which aligns the abstract concepts of normality and abnormality across the point cloud and RGB streams. Extensive experiments demonstrate that the proposed AggCLIP framework achieves state-of-the-art performance on the MVTec 3D-AD and Eyecandies benchmarks under both one-vs-rest and cross-dataset zero-shot settings, notably surpassing the strongest competitive methods by up to 2.9% in pixel-level AUPRO.

cs.CV

新论文 · v1arXiv:2611.24541

Fuse Before Reconstruct: A Novel Reconstruction Paradigm for Multimodal Industrial Anomaly Detection

Zhi Li, Hanyu Wang, Qiyu Chen, Fei Shen, Zhengtao Zhang

原文摘要

Industrial anomaly detection plays a pivotal role in precision manufacturing, where joint analysis of multimodal data emerges as a critical approach to enhance detection accuracy. Feature reconstruction, a dominant paradigm for such joint analysis, is primarily implemented through intra-modal and cross-modal methods, both of which are plagued by significant limitations. Intra-modal reconstruction neglects complementary cross-modal information, while its cross-modal counterpart is prone to projection errors and ambiguous one-to-many mappings (e.g., same textures mapping to multiple geometries), leading to detection failures. To address these limitations, we propose a novel feature reconstruction paradigm, Fuse Before Reconstruct (FBR), which first fuses multimodal features and then performs reconstruction for anomaly detection. Firstly, we propose a bidirectional cross-modal attention mechanism to achieve deep feature complementarity between image and point cloud features, mitigating projection bias through channel alignment and dynamic spatial prior integration. Besides, a learnable pixel-wise gated fusion strategy is adopted to preserve multi-modal structural consistency while optimizing local feature combination. Finally, dual-path collaborative reconstruction is implemented to recover original modal features from the fused representations, where cross-modal consistency constraints effectively eliminate mapping ambiguities. Extensive experiments show that FBR achieves 98.3% I-AUROC on MVTec 3D-AD and 92.6% I-AUROC on Eyecandies, surpassing previous state-of-the-art methods by +2.9% and +4.1%, respectively.

cs.CV

版本更新 · v2arXiv:2610.35385

EmbodiedLab: Beyond Manipulation, Benchmarking Embodied Agents That Act to Know

Hongcheng Gao, Zelin Zheng, Alex Zhou, Jay Zhu, Shijia Ge, Yazhe Wang, Xiping Li, Jianshu Zeng 等 21 位作者

原文摘要

Physical action not only changes the state of an environment but is also the way to know its hidden properties: what cannot be seen must be inferred from the feedback to the change. As embodied agents advance from executing procedures to pursuing goals, success depends on properties neither given nor visible, so they must act to know. Existing benchmarks separate changing from knowing: manipulation benchmarks give the target state, so nothing has to be found out, whereas exploration and discovery benchmarks let the agent find out by looking or querying, so nothing physical has to be changed. Quantitative science experiments offer a natural testbed: the properties sought are new in every instance, appear only in instrument readings under conditions the agent sets up, and can be checked exactly. We introduce EmbodiedLab, a simulated laboratory of 35 such experiments. Given only a goal, the agent must decide what to measure, operate the apparatus with a robot arm, read instruments through cameras and judge when its readings suffice. Its 30 determination tasks specify what to find out, whereas its 5 optimization tasks ask only for the best result the apparatus allows. With native control, the strongest of seven frontier agents succeeds on 41.4% of the tasks. Ablations reveal two bottlenecks: contact-rich manipulation, where ground-truth depth and motion planning change the success rate by at most 10 points, whereas executing every operation for the agent on five of the tasks raises it from at most 20% to at least 70%, so what an agent can know is bounded by how reliably it acts; and closed-loop experimental rigor, where with manipulation supplied, agents fail to gather sufficient readings or read instruments accurately, and most optimization runs stop at one design. Beyond evaluation, EmbodiedLab offers a resamplable environment with verifiable rewards for training embodied agents.

cs.RO

版本更新 · v2arXiv:2603.20422

PEARL: Personalized Streaming Video Understanding with Runtime-Defined Concepts

Yuanhong Zheng, Ruichuan An, Xiaopeng Lin, Yuxing Liu, Sihan Yang, Huanyu Zhang, Haodong Li, Qintong Zhang 等 13 位作者

原文摘要

Unlike humans, who learn new concepts from a continuous visual stream, existing personalized vision-language models (VLMs) internalize a predefined concept set during training and operate on static images or offline videos. They cannot register a concept that a user defines in the middle of a stream, nor recall the earlier moments in which it appeared. We formalize this task as Personalized Streaming Video Understanding (PSVU) and introduce PSVU-Bench, the first benchmark for this task, which annotates 132 videos with timestamped questions. Its Concept-Definition, Real-Time and Past-Time QA ask a model to register concepts at arbitrary timestamps, recognize their current state, including user-defined actions, and retrieve their historical evidence. We propose PEARL, a strong plug-and-play baseline that stores user-defined concepts apart from archived video clips and rewrites each query with concept descriptions to retrieve the relevant history. Grounding a named concept in video is a skill shared by all concepts. We therefore construct PSVU-10K, a training set of verified concept questions on public internet videos, and train the retriever and answering VLM of PEARL on it once. Concepts defined at test time still require no training. Applied to three architectures without training, PEARL improves Real-Time QA by 15.05% and Past-Time QA by 12.60% on average over the corresponding offline models, and all three surpass existing online video models on both QA types. Training further improves PEARL on Qwen3-VL-8B by 8.66% on Real-Time QA and 9.39% on Past-Time QA, with gains in every category. Code, data and models are available at https://github.com/Yuanhong-Zheng/PEARL.

cs.CV · cs.AI · cs.IR

新论文 · v1arXiv:2611.23926

GTI: Grounding Task Intent Throughout Long-Horizon VLA Manipulation

Tianqi Gong, Haochen Niu, Yusheng Zhao, Lantao Zhang, Haoyu Liu, Rendong Ying, Peilin Liu

原文摘要

Vision-language-action (VLA) models generate local actions from visual observations, language, and robot state, but long-horizon manipulation requires the global task intent to remain grounded in the currently executable context. We propose GTI (Grounding Task Intent), which uses the current observation and execution history to localize the active subtask and its execution phase. For the same local subtask, GTI generates two phase-dependent visual targets: an interaction/contact state and a local completion state. The local language instruction and the corresponding phase-specific visual target jointly condition the VLA, specifying both what to do now and what state should be reached at the current phase. During execution, anomaly monitoring and local recovery re-establish an intent-consistent execution context through rollback, re-observation, and phase re-evaluation after persistent deviation. We evaluate GTI on RoboCasa365, VLA-Arena Horizon-Long, and real-world PiPER tasks. Relative to matched Global-FT baselines, GR00T success on seen composite RoboCasa365 tasks increases from 14.6%/12.1% to 31.8%/20.4% under ID/OOD evaluation. On VLA-Arena, GTI+$π_{0.5}$ reaches 47.6% overall success versus 45.2% for an adapted ForeAct baseline, with gains of 5.2/6.0 percentage points on the harder L1/L2 tasks. In the three real-robot task experiments, mean full-task success increases from 21.7% to 59.2%. The results indicate that persistent semantic intent localization and phase-conditioned visual grounding improve multi-stage execution, while failure-triggered local recovery provides additional deployment-time support.

cs.RO

版本更新 · v2arXiv:2605.10228

FLARE: Full-Modality Long-Video Audiovisual Retrieval Benchmark with User-Simulated Queries

Qijie You, Hao Liang, Mingrui Chen, Bohan Zeng, Meiyi Qiang, Zhenhao Wong, Wentao Zhang

原文摘要

As video becomes increasingly central to information dissemination and multimodal large language models (MLLMs) continue to advance, evaluating video retrieval has become increasingly important. In realistic search scenarios, this requires matching short user queries to long-form content using both visual and auditory evidence. Yet existing retrieval benchmarks are still dominated by short clips, single modalities, and caption-based evaluation. We introduce FLARE, a full-modality long-video audiovisual retrieval benchmark with user-simulated queries. Built from 399 carefully screened Video-MME videos (10--60 min, 225.4 h) to ensure source quality and diversity, FLARE contains 87,697 clips annotated with vision, audio, and unified audiovisual captions, together with 274,933 user-style queries. Cross-modal queries are further filtered by a hard bimodal constraint, requiring retrieval to fail under either modality alone but succeed when both are combined. FLARE evaluates models under two regimes, caption-based and query-based retrieval, across vision, audio, and unified audiovisual settings. Experiments with 15 representative retrievers show that user-style queries substantially change model behavior, strong caption-based performance does not always transfer to query-based retrieval, and audio--language alignment remains a key bottleneck for unified audiovisual retrieval. Our code and data are released at https://flarebench.github.io/

cs.MM

新论文 · v1arXiv:2611.23746

ZERITH V2.5: From Predictive Power to Policy Learnability

Chenxi Han, Ruby, Epoch, Orion, Luke, Bin, Dusk, Lyra 等 21 位作者

原文摘要

When humans learn to manipulate an unfamiliar object, they may already be able to anticipate how its state will change while still being unable to control it skillfully. This gap between prediction and control motivates us to study how the priors acquired by video generation models can be used effectively by an action policy. We present ZERITH V2.5, a video-action model (VAM) in which the action policy reads intermediate features of a video generation model, and use it to study, under fixed demonstrations and a fixed total number of updates, what kind of predictive training improves subsequent action learning. We find that the way video predictive training is supervised affects subsequent action learning. In our experiments, adapting at many noise levels yields a large gain, while adapting only at the noise level used at deployment yields none. Weight-transfer experiments further show that the gain stays with the adapted weights when they remain in the video generation model and disappears when they are used only to initialize the action policy. Initial loss and gradient measurements at the start of joint training indicate that target-domain predictive training gives the video model a better starting state, making subsequent joint learning of prediction and action easier. Under matched data and training budgets, predictive adaptation improves control. We evaluate the complete policy on LIBERO, LIBERO-Plus, Meta-World MT50, and real-robot tasks.

cs.RO

版本更新 · v3arXiv:2610.26825

MemT-1.0: Towards Memory-Native Evolution

Zhiyu Li, Zeyu Zhang, Linyun Liu, Yiwei Bu, Chenyang Xi, Xichong Zhang, Yi Yu, Yu Yu 等 74 位作者

原文摘要

Memory-native evolution requires making memory an intrinsic capability of foundation models, enabling them to continuously accumulate and utilize information. To this end, we introduce MemT-1.0, a new architecture of memory foundation models that internalize memory as part of model computation. Building on Metis, MemT-1.0 employs Gated DeltaNet-2 with a stable token-level memory update mechanism within each turn. We further introduce Mixture-of-Storage to scale native memory capacity. Following a deliberate backbone preparation, we develop a two-stage training pipeline consisting of continued pre-training and post-training, supported by 34B tokens of training data and efficient training and inference infrastructure. We release a model family comprising MemT-1.0-Nano, MemT-1.0-Flash, and MemT-1.0-Pro, and conduct a comprehensive evaluation of memory and general capabilities. The results show that they outperform the parametric memory baselines on average and remain competitive with retrieval-augmented methods, while largely preserving general capabilities. At a history length of 128K tokens, MemT-1.0-Nano achieves approximately $1.86\times$ end-to-end inference speedup and a $66\times$ reduction in persistent storage relative to full context KV cache. By internalizing retrieved procedural experience, MemT-1.0 improves task success on agentic tasks, providing a practical step toward memory-native evolution.

cs.CL · cs.LG

新论文 · v1arXiv:2611.22356

SR-VLA: Stall-Reduced VLA Execution via Predicted-Input Lookahead and Selective KV Cache Repair

Tao Zhang, Yi Zhong, Jinhao Ruan, Yipeng Gao, Linfeng Zhong, Youming Yang, Yuan Wang

原文摘要

Vision-language-action (VLA) policies incur control stalls when inference blocks the next action chunk. Asynchronous early inference can hide latency, but its results may be inconsistent with the arriving actual inputs. Unconditional reuse risks acting on mismatched inputs, whereas full recomputation discards useful precomputation. We propose SR-VLA, combining predicted-input lookahead with selective KV cache repair. During current chunk execution, a lightweight action-conditioned predictor generates next-boundary visual observations and proprioceptive state. These predictions, together with the task instruction, drive the frozen VLA to precompute a speculative KV cache and action chunk. After real inputs are available to VLA, predicted--real differences and task conditions determine whether to accept the action chunk or trigger selective KV repair. On LIBERO, SR-VLA achieves $2.67\times$ and $1.27\times$ query speedups over the respective synchronous $π_{0.5}$ and OpenVLA-OFT baselines, while maintaining competitive average task success rates. Under target-motion perturbation, success remains baseline-comparable, with $1.52\times$--$1.93\times$ query speedup.

cs.RO · eess.SY

新论文 · v1arXiv:2611.22052

RiDEx: Reward-Informed Directional Exploration for Diffusion Reinforcement Learning

Song Yan, Chenfeng Wang, Wei Zhai, Xingchen Wang, Tao Zhang, Yawen Shao, Shi Pan, Min Li 等 9 位作者

原文摘要

Reward-based diffusion post-training updates model parameters from task rewards. While existing approaches improve sampling and policy optimization, they still rely largely on isotropic random perturbations in a high-dimensional latent space for exploration. Such variation often spends the perturbation budget on changes with little influence on the target reward, making exploration inefficient. We study where diffusion models should explore and how to learn from that exploration. We introduce Reward-Informed Directional Exploration (RiDEx), which concentrates local exploration in reward-relevant subspaces of the denoising velocity field. Conditional residuals or differentiable reward gradients propose the subspace, and equal-and-opposite perturbations at the same generation state measure its terminal reward response. The reward difference supplies signed local supervision in the explored coordinates, while all generated images contribute to global DiffusionNFT learning. RiDEx thereby couples exploration-space selection with feedback about the changes it produces. On Stable Diffusion 3.5 Medium, RiDEx attains the highest peak rewards among the evaluated methods on four alignment and preference tasks. Its gradient-direction instantiation achieves 15x and 29x target-attainment speedups over DiffusionNFT on HPSv2 and PickScore, measured in H200 GPU hours. Controlled experiments confirm that reward-aligned directions elicit stronger responses at equal perturbation magnitude, and that applying their measured credit improves curve-average reward across all four tasks.

cs.LG · cs.CV

版本更新 · v3arXiv:2609.38166

LeapQuant: Efficient Linear Attention with Accurate Recurrent State Quantization

Yi Pan, Haocheng Xi, Kan Zhu, Xingyang Li, Yibo Wu, Mayank Mishra, Hongtao Zhang, William X. Zheng 等 13 位作者

原文摘要

Recent LLMs increasingly adopt hybrid designs that replace standard attention with linear attention, such as Gated DeltaNet (GDN) and Kimi Delta Attention (KDA). Although they compress the context into a fixed-size recurrent state and substantially reduce the cost of long-context processing, repeatedly reading and updating that state remains a major inference bottleneck. Quantization offers a natural way to reduce this cost, but can significantly degrade model quality, due to the accumulation of rounding errors and the presence of outlier rows and columns in the state. To address these challenges, we propose LeapQuant, a training-free method that achieves near-lossless performance under 8-bit recurrent-state quantization. First, to mitigate error accumulation, we propose per-window quantization, which leaps over a window of tokens and quantizes the state only once at its end. Within a window, outputs are computed from the fixed low-bit state together with high-precision buffered updates. Second, to reduce the error introduced by each quantization, LeapQuant retains the state's largest outliers as a few high-precision Compensator Tokens, which share the update path of real tokens. We then smooth the remaining residual before quantization to further reduce the error. Comprehensive experiments across the Qwen, Kimi, and GLM model families show that LeapQuant substantially reduces memory and compute costs during inference. With accuracy comparable to the FP32 baseline, it achieves average speedups of 2.05--3.70$\times$ at the kernel level and 1.47$\times$ for end-to-end inference on NVIDIA B200, RTX PRO 6000, and RTX 5090 GPUs.

cs.LG · cs.AI

新论文 · v1arXiv:2611.20922

Experience Rot: Experienced Agents Fail Where Novices Succeed

Zhen Xu, Hang Yin, Zitao Zhang, Kevin Zhao, Xuefeng Liu

原文摘要

Agents are increasingly built to learn from experience: they keep conversation histories, memory files, retrieved notes and skills, and reuse what worked before. We show that this experience turns against them when their tools change silently. We test a range of models with tools defined as in the Model Context Protocol (MCP). After a silent change, experienced agents fail on most tasks, while agents new to the task mostly succeed. We call this failure Experience Rot. Experienced agents copy their earlier call instead of checking the current interface, even when the tool definition shows the change. Then they report success. When the documentation is incomplete, experience is what makes agents succeed. After a change, the same experience leaves them worse off than no experience at all. The failure appears in every form of memory we test and grows with the amount of experience. Simple remedies fall short: announcing the update barely helps, and pointing the agent to the documentation fails whenever the documentation does not spell out the answer.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.19765

M-CROSS: A Multisensor Simulation Dataset for SLAM across Water--Land Transitions

Yan Dong, Tianyi Wu, Junyu Yang, Shaoqiang Qiu, Zefeng Yan, Tao Zhang, Bin Han

原文摘要

Water--land cross-domain robots require continuous localization across land, transition and underwater phases, where sensor characteristics and availability change abruptly. Existing datasets and conventional metrics are not designed to capture transition-induced degradation and failures. We present \textbf{M-CROSS}, to the best of our knowledge, the first multisensor simulation dataset and benchmark for SLAM across continuous water--land transitions. It provides synchronized LiDAR, sonar, stereo camera, IMU, DVL, and depth measurements; 6DoF pose ground truth; sensor-wise medium-state annotations; and medium-dependent sensor degradation models. We establish a transition-aware protocol comprising the Trajectory Completion Rate (TCR), Piecewise Absolute Pose Error (PW-APE), Transition End Error (TEE), and Transition Survival Period (TSP), which measure trajectory completion, single-domain accuracy, transition-induced error, and localization survival under prolonged partial immersion, respectively. Experiments with 6 representative SLAM systems show that existing methods fail to maintain both localization accuracy and trajectory continuity across the full set of water--land transitions, exhibiting transition-induced drift, trajectory discontinuities, and tracking failures. M-CROSS is publicly available https://anonymous.4open.science/r/MCross-Dataset

cs.RO

版本更新 · v3arXiv:2601.08676

Advancing ESG Intelligence: An Expert-level Agent and Comprehensive Benchmark for Sustainable Finance

Yilei Zhao, Wentao Zhang, Lei Xiao, Yandan Zheng, Mengpu Liu, Wei Yang Bryan Lim

原文摘要

Environmental, social, and governance (ESG) criteria are essential for evaluating corporate sustainability and ethical performance. However, professional ESG analysis is hindered by data fragmentation across unstructured sources, and existing large language models (LLMs) often struggle with the complex, multi-step workflows required for rigorous auditing. To address these limitations, we introduce ESGAgent, a hierarchical multi-agent system empowered by a specialized toolset, including retrieval augmentation, web search and domain-specific functions, to generate in-depth ESG analysis. Complementing this agentic system, we present a comprehensive three-level benchmark derived from 310 corporate sustainability reports, designed to evaluate capabilities ranging from atomic common-sense questions to the generation of integrated, in-depth analysis. Empirical evaluations demonstrate that ESGAgent outperforms state-of-the-art closed-source LLMs with an average accuracy of 84.15% on atomic question-answering tasks, and excels in professional report generation by integrating rich charts and verifiable references. These findings confirm the diagnostic value of our benchmark, establishing it as a vital testbed for assessing general and advanced agentic capabilities in high-stakes vertical domains.

cs.AI

新论文 · v1arXiv:2611.16467

SS-HSDP: Efficient and Flexible Sharded Data Parallelism via Runtime-Update Decoupling

Chendong Xiang, Pengle Zhang, Jinliang Li, Zichen Liang, Jintao Zhang, Fan Bao, Jun Zhu, Jianfei Chen

原文摘要

Sharded data parallelism enables memory-efficient training of large models. While efficient training kernels favor packed, kernel-specific data layouts, optimizers may require variant data structures as optimization units. Existing sharded data-parallel frameworks couple these two choices, limiting runtime efficiency and optimization flexibility. To address this, we present \textbf{\our}, a training framework that separates runtime parameters from update parameters, allowing computation layouts and optimization units to be organized independently. Through shared data views and hierarchical sharding, \our supports this separation while avoiding redundant data movement. This design supports a broad range of optimizers, including matrix-based methods such as Muon, and performance optimizations such as fully zero-copy execution. Experiments on video and language model workloads with MXFP8 training show that \our achieves higher Model FLOPs Utilization (MFU) than FSDP2, veScale-FSDP, and Megatron-FSDP. At 128 GPUs, MFU improves by up to 21.7\% over the fastest evaluated baseline, with comparable or lower peak memory usage.

cs.DC

新论文 · v1arXiv:2611.16377

Universal optimality on finite Lee tori

Tao Zhang

原文摘要

We study subsets of finite Lee tori that minimize pair energy among sets of the same cardinality for every discretely completely monotone potential. At half occupancy on a product of even cycles, the two checkerboards minimize both total pair energy and maximum individual energy. They are the only minimizers for every nonconstant classical completely monotone potential on $(0,\infty)$, proving the checkerboard conjecture of Bouman, Draisma, and van Leeuwaarden for this class. At the independence number of the distance one graph on $\Z_q^n$, universally optimal sets exist exactly when $q$ is even, $q=3$, $n=1$, or $(q,n)=(5,2)$. In these cases they are precisely the maximum independent sets and form a single Lee isometry class. At density $1/5$ in $\Z_5^n$, $n\ge2$, universal optimality is characterized by a weighted count of pairs at distances one and two. We prove these results using integer box occupancy bounds that strengthen Fourier linear programming and, in the quinary case, Fourier certificates with a common zero set.

math.CO · cs.IT

版本更新 · v3arXiv:2509.17100

The SAGES Critical View of Safety Challenge: A Global Benchmark for AI-Assisted Surgical Quality Assessment

Deepak Alapatt, Jennifer Eckhoff, Zhiliang Lyu, Yutong Ban, Jean-Paul Mazellier, Sarah Choksi, Kunyi Yang, Po-Hsing Chiang 等 41 位作者

原文摘要

Advances in artificial intelligence (AI) for surgical quality assessment promise to democratize access to expertise, with applications in training, guidance, and accreditation. This study presents the SAGES Critical View of Safety (CVS) Challenge, the first AI competition organized by a surgical society, using the CVS in laparoscopic cholecystectomy, a universally recommended yet inconsistently performed safety step, as an exemplar of surgical quality assessment. A global collaboration across 54 institutions in 24 countries engaged hundreds of clinicians and engineers to curate 1,000 videos annotated by 20 surgical experts according to a consensus-validated protocol. The challenge addressed key barriers to real-world deployment in surgery, including achieving high performance, capturing uncertainty in subjective assessment, and ensuring robustness to clinical variability. To enable this scale of effort, we developed EndoGlacier, a framework for managing large, heterogeneous surgical video and multi-annotator workflows. Thirteen international teams participated, achieving up to a 17% relative gain in assessment performance, over 62% reduction in calibration error, and a 17% relative improvement in robustness over the state-of-the-art. Analysis of results highlighted methodological trends linked to model performance, providing guidance for future research toward robust, clinically deployable AI for surgical quality assessment.

cs.CV

新论文 · v1arXiv:2611.14997

Same Task, Different Statement: When and How Reformulation Helps LLM Problem Solving

Wentao Zhang, Xiaoyu Zhang, Chengwei Liu, Yutao Hu, Tao Zhang, Junsheng Wu, Hongping Gan, Yang Liu

原文摘要

When people face a hard problem, they often restate it in a clearer way before solving it. We study whether the same habit helps large language models by rewriting the statement of a problem. A legal rewrite must keep the meaning, the constraints, and the correct answer unchanged. We test three kinds of rewrites. Decomposition makes hidden subgoals explicit. Typed Reformulation casts the task into a more suitable solving frame. Concept Reduction removes surface detail and exposes the task structure. We evaluate them on eight math, logic, and code benchmarks with eight models, all under the same chain-of-thought solving prompt. The effect is large but depends on the problem type. An oracle that picks the best rewrite after the fact raises accuracy by +27.6% on average. Yet no single rewrite helps every problem, and a mismatched rewrite can hurt. Still, the effect follows clear patterns. Each kind of rewrite fits certain problem types. A simple fixed rule learned from one set of problems transfers part of the gain to new problems. It gains +11.0% on the problems it chooses to rewrite, while blind rewriting loses 5.6% on a similar share of problems. Rewriting is therefore not a universal way to solve problems. The question is when and how to rewrite.

cs.AI

新论文 · v1arXiv:2611.14033

Few-Step Streaming Video Generation via Condition Shifting

Deyuan Liu, Jungang Li, Kaiwen Zheng, Kai Jiang, Jintao Chen, Yaole Wang, Ming Lin, Tao Lin 等 11 位作者

原文摘要

Few-step sampling reduces the cost of streaming video generation. Distribution Matching Distillation (DMD) supervises few-step generators with a separate fake-score network, which adds training memory and computation and must track the evolving generator. We extend APEX to streaming video by using condition shifting to estimate the velocity field of the generated distribution within the generator. The original text condition selects generation, while a shifted condition selects fake estimation using the same parameters. This removes the independent fake network while retaining S2-Avatar's frozen reference teacher, Self-Replay Forcing, and perceptual regularization. We evaluate APEX on 180-second videos using three backbone evaluations per block and a one-step spatial Refiner, with a 70-case human study and a 320-case StreamAV-Bench comparison. Compared with S2-Avatar, APEX achieves higher video instruction fulfillment on the Progressive track (3.856 versus 3.831) and higher visual aesthetics across both tracks (0.641898 versus 0.632606). Human raters favor APEX most strongly on instruction following, with +30.0% net preference, followed by consistency and video quality at +4.3% each. Motion quality favors S2-Avatar, audiovisual synchronization is predominantly tied, and usability is unchanged. Separate measurements under matched training settings show 4.4% lower peak allocated GPU memory per GPU and 9.2% shorter mean iteration time than DMD.

cs.CV · cs.AI

新论文 · v1arXiv:2611.13855

Beyond Passive Imitation: On-Policy Mid-Training Distillation for Language Models

Ruize Zheng, Xuyang Liu, Siyu Ren, Xiaobing Tu, Yinggui Wang, Jinkui Ren, Xiantao Zhang, Linfeng Zhang

原文摘要

Continual pre-training (CPT) adapts large language models (LLMs) to specialized domains, yet its standard next-token prediction (NTP) objective treats every corpus token as a fixed target, regardless of its utility to the current model. In a controlled study on the same 1B-token mathematical corpus, we observe sharply different outcomes across base models: NTP improves Llama-3.2-3B's performance on mathematical benchmarks but substantially degrades Qwen3-1.7B-Base's performance, indicating that, under a fixed training recipe, the same training sequences can have model-dependent effects. To address this problem, we introduce On-Policy Mid-Training Distillation (OPMD), which reformulates mid-training from offline text imitation into active, query-driven knowledge acquisition. At selected anchor positions, OPMD samples trajectories from the evolving student policy and applies dense token-level reverse-KL rewards from a frozen teacher, training the model on states it actually visits rather than only on ground-truth prefixes. This on-policy objective reduces the training-generation mismatch of teacher-forced NTP. On MATH-500 and Minerva, OPMD improves the Qwen student over its base performance and outperforms the evaluated NTP baseline, which degrades on both benchmarks. OPMD integrates into CPT pipelines without added inference-time overhead and offers a model-aware alternative to passive domain adaptation.

cs.CL · cs.LG

版本更新 · v2arXiv:2610.23908

ContraSafe-VL: A Contrastive Safety Preference Dataset for Vision-Language Models

Di Wu, Yanyan Zhao, Xin Lu, Hao Yang, Taotao Zhang, Yunzhen Zhao

原文摘要

Vision-Language Models (VLMs) are increasingly deployed as general-purpose multimodal assistants, making multimodal safety alignment an important challenge. Safety depends on the interplay between textual and visual content, with visual context often playing a critical role. However, VLMs may fail to use visual information appropriately, either overlooking safety-relevant visual content or over-interpreting sensitive visual content, leading to unsafe responses or over-refusal. Existing multimodal safety alignment datasets largely follow the response-preference paradigm, which teaches models how to respond but provides limited supervision on how visual context should inform the safety judgment, making it difficult to balance safety and over-refusal. To address this limitation, we introduce \textbf{ContraSafe-VL}, a \textbf{Contra}stive \textbf{Safe}ty preference dataset for \textbf{V}ision-\textbf{L}anguage Models. For each fixed textual query, ContraSafe-VL constructs paired visual contexts that result in multimodal queries with different safety labels, providing direct supervision for how visual context should inform multimodal safety judgments. It further provides response preference pairs based on harmlessness and helpfulness to encourage models to generate responses that are both safe and helpful. Experiments across three VLM architectures and multiple benchmarks demonstrate that training with ContraSafe-VL improves cross-modal safety recognition and achieves a better safety--over-refusal trade-off while preserving general multimodal capabilities. The dataset is available at https://github.com/pikepokenew/ContraSafe-VL

cs.CV

新论文 · v1arXiv:2611.10586

M3FT: Micro-Macro-Micro Reasoning for Multi-Agent Cryptocurrency Perpetual Futures Trading

Jiashun Wang, Beile Wang, Guangyan Gan, Wentao Zhang, Siguang Li, Shuo Sun, Yunbo Lu, Lin William Cong

原文摘要

Cryptocurrency perpetual futures form a large-scale online financial market in which automated agents manage interacting asset positions under changing conditions. Policies relying solely on micro-level numerical states may overlook cross-asset relationships and become vulnerable to regime shifts, while macro-level linguistic judgments alone cannot ensure precise measurement or optimize positions under trading costs and risk constraints. Connecting market interpretation to coordinated multi-asset decisions remains challenging. We propose M3FT, a micro-macro-micro framework for multi-agent perpetual futures trading. The first micro stage constructs time-aligned numerical states from exchange and account observations through deterministic processing. The macro stage transforms these measured states into explicit market interpretations without adding external market information. Four asset-specific teams, each comprising a Market Interpreter, an Asset Communicator, and a Risk Reviewer, refine these interpretations through experience retrieval, cross-asset communication, and review of joint-position risks. The final micro stage combines encoded linguistic representations with the original numerical states to learn continuous target positions. We implement this framework with a language-conditioned multi-agent policy: asset actors use global and private numerical states together with linguistic context, while centralized critics evaluate joint actions using portfolio-wide information and shared-account rewards. Experiments on real cryptocurrency perpetual futures data, including comparisons across language models and policy-learning algorithms and component ablations, show that the best-performing M3FT configuration achieves a higher mean test return than the evaluated direct LLM, numerical RL, and rule-based baselines.

cs.MA

版本更新 · v2arXiv:2607.02512

Program-as-Weights: A Programming Paradigm for Fuzzy Functions

Wentao Zhang, Liliana Hotsko, Woojeong Kim, Pengyu Nie, Stuart Shieber, Yuntian Deng

原文摘要

Many everyday programming tasks resist clean rule-based implementation, such as alerting on important log lines, repairing malformed JSON, or ranking search results by intent, and are increasingly outsourced to large language model APIs at the cost of locality, reproducibility, and price. We propose fuzzy-function programming: compiling such a function from a natural-language specification into a compact, locally executable neural artifact. We instantiate this paradigm with Program-as-Weights (PAW), in which a 4B compiler trained on FuzzyBench, a 10M-example dataset we release, emits parameter-efficient adapters for a frozen, lightweight interpreter. A 0.6B Qwen3 interpreter executing PAW programs matches the performance of direct prompting of Qwen3-32B, while using roughly one fiftieth of the inference memory and running at 30 tokens/s on a MacBook M3. PAW reframes the foundation model from a per-input problem solver into a tool builder: invoked once per function definition, it produces a small reusable artifact whose subsequent calls per function application are cheap and offline.

cs.LG · cs.AI · cs.CL

新论文 · v1arXiv:2611.10463

LongCapHarness: Capability-Guided Auto-Harnessing for Long-Context Reasoning

Tao Zhang, Yuhang He

原文摘要

Existing methods enhancing an LLM's long-context reasoning capability often rely on data synthesis and follow-up LLM post-training. However, the data synthesis strategy and the post-training recipes are empirically driven and computationally intensive. In this work, we propose a new framework, dubbed LongCapHarness, to instead harness the underlying capabilities for long-context reasoning. The harnessing process recursively revises and updates capability primitive implementation, capability state and control logic through recursive self-improvement on the development set. With neither data synthesis nor LLM post-training, LongCapHarness is applied to Llama3-8B and transferred to Qwen2.5-3B. LongCapHarness substantially outperforms recent data-synthesis-based baselines on HELMET and RULER benchmarks; its estimated development cost is less than 50% of the estimated training costs of NExtLong and PolicyLong. We hope these results motivate further research into harnessing the latent capabilities of LLMs to solve tasks beyond their explicit training.

cs.CL

新论文 · v1arXiv:2611.10329

SIRIS: Can LLMs Tell When Simulation Errors Matter in AI-Driven Systems Research

Jinwoo Hwang, Sukmin Cho, Tao Zhang, Soheil Abbasloo, Jongse Park, Peng Cheng

原文摘要

LLM agents increasingly drive systems research and rely on simulators to evaluate candidate designs at scale. Simulators are imperfect by design: their modeling limitations can distort predictions yet leave the recommended design correct, or change which design should be selected. We introduce Siris to test whether LLMs can tell these cases apart, judging from their systems understanding when a limitation invalidates the simulator's recommendation. By perturbing a simulator, Siris constructs workload pairs with the same candidates, limitation, and simulator recommendation. The perturbation changes predictions on both workloads, yet a reference simulator supports the recommendation on only one. Evaluating thirteen frontier LLMs on LLM serving design problems, we find that identifying a limitation does not ensure a correct choice, and even the strongest models sometimes abandon valid recommendations. Self-correction through a review turn helps these models restore valid choices but introduces new errors in several others. These findings motivate evaluating systems understanding through design decisions under imperfect simulators.

cs.AI

新论文 · v1arXiv:2611.09340

Horizon-Coupled Obstacle Guidance for Relative-Action Diffusion Policies

Guangyuan Wang, Haishan Xia, Zitao Zhang, Kejia Chen, Yuanchen Sun, Zhenguo Sun, Donghao Zhang, Yu Zhang 等 9 位作者

原文摘要

Real-world robot deployment requires adapting learned policies to unseen obstacles without collecting new demonstrations or retraining. Existing policy-level guidance often applies obstacle corrections independently at individual horizon steps, which may address local obstacle violations but fail to coordinate the action sequence into a complete, task-directed detour. How obstacle responses should be propagated across the action horizon to coordinate complete, task-directed avoidance motion remains underexplored. We introduce an inferencetime obstacle-guidance framework for pretrained relative-action diffusion policies. It rolls out cumulative relative actions into a workspace trajectory, evaluates obstacle interactions along the trajectory with an SDF-based obstacle model, and propagates each obstacle response to all contributing action increments during denoising. Real-robot experiments across seven obstacleconstrained scenarios spanning four manipulation tasks show that our method achieves 105/140 (75.0%) collision-free task successes, compared with 8/140 (5.7%) for absolute-action pointwise guidance. Ablations show that the relative-action horizoncoupled formulation provides the larger observed improvement across both obstacle models.

cs.RO

新论文 · v1arXiv:2611.08905

From Procedures to Permissions: Learning Precise Subagent Authorization from Agent Skills

Wenkai Yu, Ning Ding, Wentao Zhang, Yehui Tang

原文摘要

As language agents tackle increasingly complex tasks, they increasingly rely on multiple agents to coordinate specialized work, requiring decisions not only about what each subagent should do but also which tools and workspace resources it should access. Existing agent systems provide mechanisms for controlling tools and execution environments, but do not directly specify the resources required by each delegated subtask. Moreover, task outcomes and execution trajectories alone provide limited supervision for determining whether a subagent has received appropriate permissions. We observe that Agent Skills encode procedural knowledge describing the operations, tools, and files required at each step, providing a natural source of supervision for authorization. Building on this observation, we synthesize tasks and step-by-step procedures from Skills, execute them through subagent delegation in interactive environments, and use verified trajectories to construct training data for task execution and authorization. On ClawArena-Team, our method improves tool permission precision by 13.4 percentage points and workspace permission precision by 24.1 percentage points, while increasing overall task success by 11.6 percentage points for Qwen3.5-35B-A3B. On Terminal-Bench 2.1 and SkillsBench, our method remains competitive with existing agent data synthesis approaches. These results show that the procedural knowledge embedded in Agent Skills provides a source of supervision for precise subagent orchestration that task outcomes alone cannot provide.

cs.CL