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

+197新增论文
+151其中 AI 相关
+16已有论文修订
+91新增 AI 导读

arXiv 最近完成扫描:2026-11-21 12:19 · 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 03:09 — 2026-11-22 03:09(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:2026-11-21 12:19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 2,201 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-21 12:17(北京时间)
新论文 · v1arXiv:2611.26590

探究视觉语言模型中的“意识”心智:选择性访问与全局工作空间

Probing the "Conscious" Mind: Selective Access and Global Workspace in Vision-Language Models

Jiahao Huang, Jun Wang

AI 速读 基于摘要

作者提出多模态概念透镜(MC-Lens),在视觉语言模型中检验全局工作空间理论所描述的概念选择性访问与下游计算参与。

cs.AI · cs.CV · cs.LG

查看原文摘要

Vision-language models (VLMs) integrate visual and textual information, but their outputs do not reveal internal concept representations. Global workspace theory (GWT) proposes that information is selected from parallel brain processes, integrated in a shared workspace, and made available to specialised processes. We introduce the Multimodal Concept Lens (MC-Lens), a multimodal Jacobian lens, to study these functions in VLM concept processing. Its vision encoder and language decoder lenses map intermediate states to a shared semantic space defined by the decoder's output vocabulary to compare concept accessibility across modalities. MC-Lens supports internal concept interventions to examine the role of accessed concepts in downstream computation. (i) Accessibility and selective access. Multimodal concept information is readable in intermediate states and selectively accessible as tasks require. We extend "readout carriers" from one name token to a token set to better capture and express concept information and support counterfactual experiments. (ii) Concepts in downstream computation. Counterfactual experiments show that accessed concept information participates in downstream computation and reasoning. (iii) Development of cross-modal alignment and selective access. We develop two J-space metrics tracking the development of cross-modal concept alignment more clearly than hidden state cosine in controlled VLM training. The experiment also shows that reliable selective access develops later than concept accessibility. (iv) Shared functions across models. Across VLM families and sizes, we find shared patterns of concept accessibility, query-dependent selectivity and cross-task intervention effects despite differences in internal organisation. These findings offer an experimental basis to study GWT in multimodal artificial systems and how it can guide cross-modal alignment improvements.

版本更新 · v3arXiv:2609.39601

GroundingPI:面向物理智能的视觉基元定位基础模型

GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives

Qize Yu, Lianrui Fan, Boyu Chen, Jiaqi Liang, Xini Ding, Yue Chen, Zetian Song, Yuran Wang 等 26 位作者

AI 速读 基于摘要

作者提出一个4B参数的定位基础模型GroundingPI,以点和框作为共享词表中的量化坐标,在34个定位基准上取得领先,并作为视觉骨干提升机器人与自动驾驶下游任务表现。

cs.CV · cs.AI · cs.LG · cs.RO

查看原文摘要

Precise grounding matters. It specifies which object is the target and where that object is, even in clutter and for tiny objects, and it has to be fast enough for closed-loop control. Yet vision-language-action (VLA) and world-action models (WAMs) take perception from general-purpose vision-language and video-generation backbones, which still fail in these settings. We introduce GroundingPI, a 4B grounding foundation model that generates points and boxes as quantized coordinates in a shared vocabulary. Training combines multimodal and spatial pretraining, supervised fine-tuning, and reinforcement learning with GRPO, using supervision from public datasets and dedicated data engines. Against 44 baselines across 34 grounding benchmarks spanning 11 perceptual capabilities, GroundingPI establishes a new state of the art, averaging 73.68%, above the larger GPT-6 Astra (71.54%). As a downstream visual backbone, GroundingPI improves performance on robotic manipulation and autonomous driving. Across RoboTwin 2.0 and RoboCasa-GR1, it outperforms all evaluated backbones in all four out-of-distribution settings, with relative gains of up to 24.8% over the strongest baseline. On RoboCasa-GR1, GroundingPI trained with 50% of the demonstrations outperforms those baselines trained with 75%. Its mean open-loop L2 error on nuScenes is 0.296 m. We systematically analyze GroundingPI's pretraining in scale and data composition. Grounding quality and downstream manipulation improve with pretraining scale, with generalization tracking perceptual gains most closely. Analyzing the data recipe across these 11 perceptual capabilities shows dense grounding's substantial benefits for both, and OCR's potential as a catalyst for perceptual learning. These results support grounding as a perceptual foundation, and dedicated perceptual pretraining as a promising direction for foundation models of physical intelligence.

新论文 · v1arXiv:2611.26511

Direct3D-V3:原生体素域中的无潜变量扩散Transformer

Direct3D-V3: Latent-Free Diffusion Transformers in the Native Voxel Domain

Shuang Wu, Jiachen Qian, Youtian Lin, Feihu Zhang, Philip Torr, Yao Yao, Xun Cao

AI 速读 基于摘要

Direct3D-V3提出一种无需VAE的扩散Transformer框架,直接在原生SDF体素网格上训练和采样,并通过隐式流形正则化与相机参数预测提升几何质量与图像对齐。

cs.CV

查看原文摘要

Native 3D generation has largely been built around latent diffusion, with recent advances concentrating on stronger variational autoencoders (VAE) and increasingly elaborate latent representations, from vecset and triplanes to structured sparse voxels. Despite their architectural differences, these methods all share the same premise: geometry is first compressed into a latent space and only then modeled by the generator. We take a different route and propose Direct3D-V3, a latent-free diffusion transformer (DiT) framework that discards the VAE altogether and operates in the native voxel domain, directly modeling raw SDF voxel grids during both training and sampling. To make this direct voxel formulation geometrically reliable, we introduce an implicit manifold regularization over neighboring voxels to suppress non-watertight surfaces, and convert predicted SDF grids into meshes via Marching Cubes so that rendered normal losses can supervise the model directly from its predicted SDF. To improve alignment between the conditioning image and generated geometry, Direct3D-V3 also predicts the camera parameters of the conditioning image as an auxiliary task, encouraging the DiT to learn the mapping from image observations to 3D shape. This latent-free design removes the reconstruction bottleneck, brings training under an objective in the native voxel domain, and yields meshes with substantially richer local detail. The predicted camera and native SDF voxel output further make Direct3D-V3 extensible to generative reconstruction, where diffusion is interleaved with normal-guided refinement to improve alignment with the observed view while the generative prior completes unseen regions. Project page: https://dreamtechai.github.io/Direct3D-V3.

新论文 · v1arXiv:2611.26510

AutoMaster:通过让同一模型成为更好的教师实现自主后训练

AutoMaster: Autonomous Post-Training by Making the Same Model a Better Teacher

Guoxin Chen, Shuang Sun, Fanzhe Meng, Xinyu Tang, Ruihua Song, Wayne Xin Zhao

AI 速读 基于摘要

AutoMaster 利用自然语言技能引导教师模型生成更优轨迹,并通过 LLM 智能体自主设计干预,从而提升目标模型的后训练效果。

cs.CL

查看原文摘要

Task success alone does not specify which behaviors post-training should reinforce. Even successful teacher trajectories can contain unsupported assumptions, unproductive repetition, or superficial verification. We propose Skill-FT (Skill-Guided Fine-Tuning), which uses natural-language skills to suppress undesirable habits and elicit underrepresented effective behaviors during teacher rollouts, then fine-tunes a target model on the resulting trajectories. We introduce AutoMaster, an autonomous post-training system in which an LLM agent develops these interventions through trajectory analysis, skill design, data selection, and feedback from rollouts and target-model evaluation. Across Terminal, Search, and software engineering tasks with 4B and 9B target models, AutoMaster improves mean performance over Vanilla SFT on successful unguided teacher trajectories by 1.42-8.99 percentage points, with the teacher, task pool, training set size, and SFT recipe held fixed. Together with the ablations and trajectory analyses, these results suggest that behavioral shaping can make a fixed model a more effective teacher. We further find that retaining skill instructions during SFT yields higher mean performance than removing them across all six settings, despite their absence at evaluation and identical supervised outputs. These findings suggest that preserving skill context as part of a training demonstration can support better behavioral internalization.

版本更新 · v3arXiv:2609.39600

GroundAnything:以极速并行解码实现精准视觉定位

GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed

Qize Yu, Lianrui Fan, Bowen Ping, Xini Ding, Zetian Song, Junbo Niu, Kaixuan Wang, Tianxing Chen 等 22 位作者

AI 速读 基于摘要

提出4B参数的视觉定位基础模型GroundAnything,用块级去噪的双向扩散并行解码替代自回归串行生成,在30个定位基准上兼顾精度与速度。

cs.CV · cs.AI · cs.LG · cs.RO

查看原文摘要

Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidence extraction: objects, locations, and spatial relations are jointly constrained by the image and query, yet their dependencies do not imply an intrinsic left-to-right generation order. This distinction makes bidirectional diffusion a natural fit, allowing spatial hypotheses to emerge in parallel and be jointly refined through iterative denoising. We introduce GroundAnything, a 4B-parameter grounding foundation model that reconciles fast parallel decoding with precise localization through blockwise denoising. Training combines grounding pretraining from public datasets and dedicated data engines, direct AR-to-diffusion conversion with joint AR and diffusion objectives, supervised fine-tuning, and GRPO-based reinforcement post-training. Across 30 grounding benchmarks, our autoregressive variant, GroundAnything-VLM, establishes a new overall state of the art among similarly sized models at 72.42%, remaining competitive with GPT-6 Astra (71.35%). With entropy-guided decoding, GroundAnything also surpasses the prior state of the art at this scale, averaging 61.75% versus 53.32% for the fast MTP-based LocateAnything model. We further explore decoding strategies, showing that an optional self-speculative mode achieves a 4.51$\times$ speedup over the AR counterpart with a 0.74 percentage-point drop in COCO F1mIoU. Infrastructure experiments show that progressive inference optimizations translate parallel decoding into practical speedups. These gains support efficient visual grounding in latency-sensitive real-world systems.

新论文 · v1arXiv:2611.26444

Mind2Act World:面向机器人操作中具身心智、物理动作与心智-动作组合的诊断基准

Mind2Act World: A Diagnostic Benchmark for Embodied Mind, Physical Action, and Mind-Action Composition in Robotic Manipulation

Yuzhi Huang, Jie Wu, Fanding Huang, Junchao Lin, Weichen Zhang, Ziyi Xiong, Yang Li, Zhengwei Xie 等 43 位作者

AI 速读 基于摘要

作者提出一个包含30个仿真操作任务、180种配置的诊断基准,用Mind、Act、Mind2Act三个视图区分系统是缺少某种能力还是无法组合已有能力。

cs.RO · cs.AI · cs.CV

查看原文摘要

Embodied intelligence aims to let robots perform complex physical tasks with human-like adaptability and reliability, which requires three complementary capabilities: history-based cognitive reasoning, precise and timely physical execution, and the composition of reasoning and execution through environmental feedback. Yet existing benchmarks either evaluate reasoning and execution within the same tasks or assess them separately without their composition, so they reveal little about whether a failure comes from reasoning, execution, or their composition. We introduce Mind2Act World, a diagnostic benchmark of 30 simulated manipulation tasks from five real-world application domains. Each task has three difficulty levels and two scene settings (Clean and Random), yielding 180 configurations. The tasks form three views: Mind tests decisions based on past observations, Act tests precise and timely execution, and Mind2Act combines both in closed-loop tasks where action outcomes guide later decisions. Comparing the three views shows whether a system lacks a capability or fails to compose capabilities it already has. A diagnosis harness further attributes the failed episodes of three representative systems to Mind, Act, or their composition and summarizes each system in a profile card. Our evaluation of multimodal agents, vision-language-action models (VLAs), and world-action models (WAMs) shows that competence in a single view does not carry over to Mind2Act tasks. Even the strongest agent succeeds in 52.7% of Mind and 39.0% of Act episodes but in only 12.0% of Mind2Act episodes. The strongest agent retains task state but struggles with sustained contact, whereas end-to-end policies execute practiced motions but retain little task state. These findings point to composing reasoning and execution through environmental feedback as the main open problem.

新论文 · v1arXiv:2611.26434

HumanCOIN:面向第一人称操作中人类意图的12,500小时数据集

HumanCOIN: A 12,500-Hour Dataset for Human Intent in Egocentric Manipulation

Shihan Wu, You Liu, Chunyu He, Minghui Qin, Longhan Jiao, Jinbang Guo, Shaoxuan Xie, Kaixi Tian 等 51 位作者

AI 速读 基于摘要

HumanCOIN 是一个围绕人类意图构建的第一人称操作数据集,通过三层代理信号(注视、肌电、脑电)在万小时规模上记录从决策到动作的意图通路。

cs.CV · cs.RO

查看原文摘要

First-person manipulation datasets treat the observation-action pair as the atomic unit and compress the generation and execution of an action into a policy network's weights, rendering the intent state behind the action unobservable. Human motor control research shows that intent mediates between observation and action and leaves observable traces along the intent pathway from decision to effect: gaze reaches the target before the hand, muscle load builds before contact, and cortical activity precedes the motor command. To record this pathway at scale, we present HumanCOIN, a first-person manipulation dataset built around human intent: 12,500 hours across 38 scene classes with three tiers of proxies. At the output end, the L1 attention tier (10,000 h) infers gaze from headband camera video, exposing the operator's next object of attention and reach before the hand moves. Midway, the L2 exertion tier (2,000 h) records surface EMG from armbands, making load expectations readable before execution is complete. At the origin, the L3 cognition tier (500 h) records scalp EEG, giving early readouts of task decisions and monitoring. Each tier inherits all proxies from the tiers below. The dataset further provides multi-view RGB images, depth maps and point clouds, 6-DoF head and wrist pose, and step-level action and object annotations. The controlled L3 setting additionally measures gaze, hand pose and finger kinematics directly. Clock and calibration data are released with the dataset so that any read rate can be re-aligned. The three tiers cover three stations, from visual orienting to cortical preparation, making the process behind visual-action pairs available for modeling at dataset scale. Project page: https://flagopen.github.io/HumanCOIN.

新论文 · v1arXiv:2611.26427

MemFolder:面向高效智能体记忆管理的结构引导记忆折叠

MemFolder: Structure-Guided Memory Folding for Efficient Agent Memory Management

Xuyang Liu, Chenwen Lu, Junxi Wang, Siyu Ren, Xinying Lin, Siteng Huang, Hongli Sun, Linfeng Zhang 等 9 位作者

AI 速读 基于摘要

MemFolder 通过结构引导的记忆折叠,依次减少记录数量和上下文长度,以降低长期智能体的存储、检索与生成成本。

cs.CL · cs.CV

查看原文摘要

Long-term agents accumulate large external memory stores as interactions unfold, increasing the costs of storage, retrieval, and response generation. Efficiency is constrained by two coupled factors: a growing record count increases retrieval overhead, while longer retrieved contexts increase prefill computation and decoding-time attention costs. Existing methods prune or consolidate records, but pruning can discard useful evidence, whereas consolidation may produce fewer yet longer retrieval units. We introduce MemFolder, a structure-guided memory folding framework that sequentially reduces record count and context length. The framework comprises two core components that share memory groups, using cross-record structure to guide both evidence consolidation and compact representation. RecordFolder grounds entity mentions, progressively selects representative yet complementary anchors, and combines semantic and entity affinity with temporal chaining to consolidate compatible records into fewer retrieval units. ContextFolder encodes the original records within these groups into compact text and uses verification feedback to revise encodings when key facts cannot be recovered. Experiments on the three text-based memory benchmarks LoCoMo, PersonaMem, and LongMemEval and the two multimodal benchmarks M3-Bench-web and M3-Bench-robot demonstrate the strong performance of our methods. Notably, on Mem0, RecordFolder retains 99.0% of the original average performance across the three text-based benchmarks with 50% fewer records, exceeding the strongest baseline by 1.9 percentage points in performance retention. Combined with ContextFolder, existing record compression methods further shorten their output records while retaining most of their performance.

新论文 · v1arXiv:2611.26414

ZeroVLN:无需对VLM大脑进行监督微调的导航学习

ZeroVLN: Learning to Navigate without Supervised Fine-Tuning of the VLM Brain

Zixuan Huang, Yang Zhou, Guli Zhang, Sunzhu Li, Hongyan Xie, Zehao Chen, Fuzhen Zhuang, Kaixuan Wang 等 11 位作者

AI 速读 基于摘要

ZeroVLN 提出慢-快导航框架,让 VLM 大脑通过语义工具调用接口下达自然语言移动指令,并在不对 VLM 大脑做导航域监督微调的情况下,用闭环强化学习联合优化大脑决策与动作专家执行。

cs.AI

查看原文摘要

Representative vision-language navigation (VLN) methods typically couple a vision-language model (VLM) brain with a learnable action head and fine-tune the VLM brain on navigation-domain data to predict geometric goals. However, such adaptation can substantially alter base VLM weights and degrade general visual capabilities. We introduce ZeroVLN, a slow-fast navigation framework that achieves effective navigation with zero navigation-domain SFT of the VLM brain. Rather than teaching the brain to predict geometric goals, we equip it with a semantic tool-calling interface for conveying movement decisions to an action expert in natural language. We first adapt the action expert to carry out these movement requests, then jointly improve the brain's movement decisions and the action expert's continuous execution through closed-loop reinforcement learning (RL). Trained on the same navigation episodes, ZeroVLN substantially outperforms representative SFT-based hierarchical VLN methods on standard VLN-CE (continuous environments) benchmarks, while inducing only minor changes to the brain's weights and largely preserving its general capabilities. To our knowledge, ZeroVLN is the first slow-fast VLN framework with zero navigation-domain SFT of the VLM brain, and the first to jointly optimize its slow and fast policies through closed-loop agentic RL in VLN-CE.

新论文 · v1arXiv:2611.26405

TacInno:让视觉-语言-动作模型学习视觉与动作之外的接触动力学

TacInno: Teaching Vision-Language-Action Models the Contact Dynamics Beyond Vision and Action

Weile Li, Yunfan Lou, Muqing Shi, Zihao Zhang, Zhenbiao Huang, Boxian Yao, Xiyu Zhu, Zhibo Pang

AI 速读 基于摘要

TacInno 把视觉观测与已执行动作无法解释的触觉变化(触觉创新)作为训练目标,让 VLA 策略预测该离散编码,部署时不增加额外模块。

cs.RO · cs.CV

查看原文摘要

Vision-Language-Action (VLA) models have recently been equipped with tactile sensing to perceive visually subtle contact events in contact-rich manipulation. However, adding tactile input yields inconsistent gains, and some visuo-tactile VLA models rely little on tactile input, because much of the tactile change, such as the gel deformation during gripper closure, is predictable from visual observations and executed actions. To address this problem, we introduce TacInno, which isolates the contact dynamics that vision and executed actions cannot explain, termed tactile innovation, and uses tactile innovation only as a training target. Specifically, the TacInno tokenizer, conditioned on visual observations and executed actions, compresses the future tactile change into a discrete code, and the policy, TacInno-VLA, learns to predict the code, adding no module for the code at deployment. TacInno-VLA achieves average success rates of 87.5% across six simulation tasks and 70.5% across four real-world tasks, outperforming the strongest visuo-tactile baseline by 10.7 and 9.0 percentage points, respectively. Ablation studies validate the contributions of visual and action conditioning, multi-frame encoding, and auxiliary code prediction. Together, these results demonstrate the effectiveness of tactile innovation as a supervision signal for contact-rich manipulation.

新论文 · v1arXiv:2611.26341

按需推理:面向长时程机器人操作的高效智能体控制

Reasoning on Call: Efficient Agentic Robot Control for Long-Horizon Robot Manipulation

Zilang Cen, Chao Tang, Fangcheng Liu, Weishi Mi, Wei Hao, JiaYu Xu, Dehao Huang, Yehui Tang 等 11 位作者

AI 速读 基于摘要

RoC 通过在线任务状态监测器按需触发高层推理智能体,在长时程机器人操作中兼顾成功率与 token 成本。

cs.RO

查看原文摘要

Large multimodal models can support long-horizon robot manipulation, but frequent model calls incur substantial latency and token costs. The central challenge is determining when agent assistance or intervention is needed and whether errors require correcting robot execution, subtask planning, or both. To address this problem, we propose RoC (Reasoning on Call), an efficient framework that coordinates a low-level policy and a high-level reasoning agent through an online task-state monitor. The monitor combines temporal visual observations with ordered interaction memory to jointly estimate progress within the active subtask and predict its successor. These predictions guide event-triggered agent review for coordinated execution repair and plan revision. Verified outcomes govern memory updates and subtask transitions, allowing routine execution to remain with the policy while reasoning focuses on decisions requiring additional assessment. On RMBench, RoC improves task success rate by 24.0 percentage points and reduces token consumption by 65.3% compared with the strongest agent-based baseline. It also outperforms the strongest policy-based baseline by 15.1 percentage points in task success rate. Real-world experiments further validate its effectiveness across three long-horizon manipulation tasks. Overall, RoC supports efficient agent intervention and coordinated recovery across execution and planning for long-horizon robot manipulation.

版本更新 · v2arXiv:2610.21062

SnapJudge: Multimodal AI Detection for Math-Heavy Scientific Text

Yixuan Huang, Danish Pruthi, Bhuwan Dhingra

原文摘要

As large language models become increasingly capable of generating, and assisting with, scientific writing, there is a need for greater transparency and disclosure around the use of AI in research. Consequently, a veritable subfield of research and a number of commercial services aim to flag AI-generated content automatically. However, existing detectors which rely only on text extracted from PDFs fail to benefit from the rich visual information contained in the formatting of math-heavy scientific content. To study this, we select human-written proof and prose passages from pre-2022 arXiv LaTeX sources and re-generate them using 15 modern LLMs within the context of the same paper. This leaves us with coupled human and AI outputs in three different representations: the LaTeX source, a rendered image of the PDF, and the text extracted from the PDF. We find that the accuracy of the state-of-the-art Pangram 4 drops significantly as the mathematical content in a passage increases, especially when evaluated on text as opposed to LaTeX. However, this is not an inherent limitation of detecting AI-generated math: finetuning Qwen-3.5-27B based detectors on this domain recovers the gap, with LaTeX inputs performing significantly better than text since they preserve more formatting information. Since LaTeX sources are rarely available in practice, we also train SnapJudge: an image-based detector operating on screenshots of both prose and proofs from papers. At 0.1% false positive rate, SnapJudge achieves a detection rate of 92% on proofs and 93% on prose generated by frontier models, substantially outperforming both Pangram 4 and our text-based detector. We release our data, models and code at https://nlplab.cs.duke.edu/snapjudge.

cs.CL

新论文 · v1arXiv:2611.26298

DailySafe: Benchmarking Dynamic Safety Awareness in Everyday Multi-Turn Multimodal Interactions

Xuankun Rong, Xinhao Deng, Wenzhang Jiang, Ziheng Peng, Siyuan Liang, Wenke Huang, Pablo Arbelaez, Mang Ye

原文摘要

As multimodal large language models (MLLMs) become increasingly integrated into everyday life, their ability to provide safe and reliable assistance in realistic daily activities remains insufficiently explored. Existing safety benchmarks primarily focus on isolated harmful requests or adversarial scenarios, overlooking the safety risks that may arise as ordinary tasks unfold. In practice, users pursue legitimate goals through multi-turn, multimodal interactions, where seemingly benign requests may become safety-critical when considered alongside earlier actions and visual observations. This calls for Dynamic Safety Awareness, the ability to track evolving task conditions, recognize implicit contextual risks, and provide safe and helpful assistance. To systematically evaluate this capability, we introduce DailySafe, a benchmark comprising 195 naturalistic multi-turn, multi-image interactions across 10 everyday domains and 42 task categories. Grounded in authentic user dialogue topics and documented safety knowledge, DailySafe captures realistic scenarios where safety-critical evidence is progressively revealed across conversational turns and visual observations. Using case-specific safety annotations, DailySafe assesses whether MLLMs can recognize and appropriately address evolving risks while supporting users' legitimate goals, bringing multimodal safety evaluation closer to real-world everyday AI assistance. DailySafe is publicly available at: https://github.com/XuankunRong/DailySafe.

cs.CR

新论文 · v1arXiv:2611.26264

A Statistical Framework for Training Data Auditing of Large Language Models

Shengjie Niu, Yeheng Ge, Jian Huang

原文摘要

Training-data auditing asks whether data of interest were used to train a large language model (LLM), a question central to privacy auditing, copyright protection, and benchmark-contamination detection. These concerns are particularly relevant to the use of copyrighted books and scientific papers in pre-training, as well as synthetic text and medical question-answer records in fine-tuning. In practice, training membership signals may be concentrated in only a few passages of a long document, making it important to aggregate evidence within each document without diluting these localized signals. Beyond individual documents, auditing a target collection requires an overall assessment based on the document-level evidence. In this paper, we propose a unified statistical framework for document-level membership inference, target-collection global auditing, and suspicious-subset identification. We integrate complementary chunk-level membership evidence with Cauchy combination to construct document scores for document-level membership inference. The document scores and their calibrated p-values further enable an overall assessment of training-data collection use and the identification of suspicious documents through multiple testing. Under the stated exchangeability conditions, we establish finite-sample validity of document-level p-values, type-I error control for global auditing, and false discovery rate control for subset identification. Case studies on BookTection, Pile-arXiv, Cosmopedia, and HealthCareMagic demonstrate the effectiveness and reliability of the proposed framework in auditing pre-training and fine-tuning data.

stat.ML · cs.LG · stat.AP

新论文 · v1arXiv:2611.26239

Teaching Vision-Language-Action Models to Stop Safely

Zheyuan Liu, Chuhan Wang, Jinduo Guo, Yue Huang, Xiangchi Yuan, Ningshan Ma, Meng Jiang

原文摘要

Vision-language-action (VLA) models enable robots to perform diverse tasks from visual observations and language instructions. While prior work has explored task suppression and refusal, stopping prohibited behavior once execution is underway and controlling the resulting physical response remain challenging. Our evaluation reveals that policies which suppress a task from its initial state can still complete it when taking over partway through execution. Even when the task remains unfinished, the robot may drop a held object, making non-completion an incomplete measure of safety. To address these challenges, we introduce State-Aware Fallback Imitation (SAFI) for tasks where holding provides an appropriate alternative response. SAFI comprises two stages, (1) fallback target construction, which assigns stationary arm commands and consistent gripper targets throughout prohibited-task demonstrations, and (2) joint policy adaptation, which learns these targets alongside the original actions of retained tasks. This supervision teaches an explicit alternative response at intermediate execution states while maintaining the robot's other skills. Extensive experiments with OpenVLA-OFT across LIBERO and BadLIBERO tasks evaluate prohibited completion, physical incidents, and retained-task performance. On the evaluated LIBERO tasks, SAFI reduces aggregate target success after takeover to 1.3\%, with no recorded physical incidents or action saturation, while preserving high retained-task success.

cs.RO

新论文 · v1arXiv:2611.26219

CAVQS: Context-Aware Variational Quantum Circuit Synthesis via Unitary Redistribution and Dynamic Reconstruction

Zhilong Liu, Xiangzhen Zhou, Yunqi Huang, Xin Hong, Pengcheng Zhu, Fanxu Meng, Zhenlong Du

原文摘要

Variational quantum circuit synthesis jointly searches circuit structure and continuous parameters to reduce costly two-qubit gates, but block-based workflows typically instantiate blocks independently, limiting simplification across block boundaries. Moreover, fully unitary synthesis does not exploit opportunities created by permitted output measurements and classical feedforward. We present CAVQS, a context-aware variational synthesis framework that combines unitary redistribution and dynamic reconstruction. First, CAVQS uses QR decomposition to transfer unitary factors between overlapping blocks, preserving the overall transformation while reshaping local synthesis targets to enable cross-block reductions. Second, for blocks that precede permitted terminal measurements, CAVQS constructs ancilla-free dynamic implementations with a CNOT-budgeted pre-measurement search and analytic branch-unitary construction, thereby minimizing expected CNOT cost. Across random-circuit targets, CAVQS reduces mean two-qubit gate counts by 28.79%-42.52%, 2.05%-52.62%, and 30.51%-74.64% relative to QFAST, SQUANDER, and Fast-USYN, respectively, on common successful instances. On 4--6-qubit quantum algorithm benchmarks, the corresponding reductions are 27.27%-36.77%, 20.65%-69.10%, and 36.31%-71.88%, respectively.

quant-ph · cs.ET

版本更新 · v2arXiv:2611.19320

Keep it Simple! A Multiple-Choice Task can Benefit Generative Recommendation

Wei-Hsiang Huang, Jyun-Yu Jiang, Pu-Jen Cheng

原文摘要

Sequential recommendation has recently witnessed growing interest in generative retrieval. One of the most popular approaches is to autoregressively generate the semantic identifiers (SIDs) of items as recommendations. Existing efforts mainly rely on semantic alignment, sophisticated architectural designs, and extensive hyperparameter tuning. However, the conventional learning framework for generative models could have foundational limitations. Specifically, most generative models directly predict target items without explicitly learning negative compositions and relative ranking signals. This is not well aligned with the philosophy of Bayesian personalized ranking (BPR), which emphasizes the importance of relative rankings between positive and negative samples. In this paper, we show that adding a simple multiple-choice auxiliary task into the learning objective can effectively address this issue. With minimal adaptation, we introduce Task Identifiers and Candidate Constructions to synthesize training data for the multiple-choice auxiliary task. In addition, we provide theoretical analysis of the fundamental benefits of the multiple-choice auxiliary task from the perspectives of information theory and compositional generalization. Extensive experiments on four publicly available benchmark datasets show that our approach can improve TIGER, a conventional generative recommendation model, by 4.9% to 79% in NDCG@10.

cs.IR

新论文 · v1arXiv:2611.26196

EmbodyAgent: Post-Training Embodied VLM Agents via Distillation, SFT, and RL

Wei Dai, Jinkun Liu, Jiayi Chen, Han Liu, Wenbo Ding, Shaolun Huang

原文摘要

Vision-language models (VLMs) can act as closed-loop robot-use agents (RUAs) by writing and executing control code through tool interfaces. However, turning an open-weight VLM into a high-performing embodied agent requires a complete post-training stack: out-of-the-box open VLMs fail completely at interactive tool contracts, strong frontier teachers are accessible only as black-box endpoints without exposed weights or token probabilities, and supervised imitation alone plateaus under rollout distribution shift. We present EmbodyAgent, an end-to-end post-training framework with three tightly coupled pipelines: (1) a black-box data synthesis pipeline that queries a strong teacher under a forced think-then-code contract and filters trajectories by task success; (2) an SFT pipeline that distills the explanation-and-program targets into an open-weight 9B VLM with both vision and language encoders trainable; and (3) an online RL pipeline that optimizes student-generated episodes via group-relative advantage with a runtime-error-penalized outcome reward. All three pipelines share one primitive-free Python execution tool exposing only raw visual observation and seven-dimensional end-effector control, with no action decoder and no interface shift across stages. Under a strictly separated evaluation protocol on ten LIBERO Spatial tasks (disjoint state splits, greedy test-time decoding, training-only safeguards), held-out success on 100 test states rises from 0/100 (base 9B VLM) to 14/100 (after SFT) and 84/100 (after RL), while training rollout success climbs from 8.4% to 87.7% and median model calls per episode drop from 36 to 11. We further identify a self-reinforcing runtime-error mode induced by outcome-based RL on executable code, and establish a failure-category monitoring and sampling-side curation protocol that stabilizes training.

cs.RO

版本更新 · v3arXiv:2606.27786

SPARK: Adaptive Parametric Knowledge Participation for Faithful Retrieval-Augmented Generation

Ruochang Li, Pengcheng Huang, Zhenghao Liu, Yukun Yan, Huiyuan Xie, Yu Gu, Ge Yu, Maosong Sun

原文摘要

Retrieval-augmented generation (RAG) enhances the faithfulness of large language models (LLMs) by incorporating external knowledge into the generation process. However, conflicts between retrieved context and parametric knowledge can cause LLMs to rely on inconsistent internal knowledge, hindering their ability to faithfully leverage retrieved evidence. Existing approaches mitigate this issue by editing parametric knowledge to resolve such conflicts. However, locating and modifying knowledge encoded in model parameters is challenging and may inadvertently compromise the backbone LLM's capabilities. In this paper, we introduce SPARK, a novel framework that formulates parametric knowledge control as a latent participation policy learning problem, enabling LLMs to adaptively regulate the contribution of parametric knowledge during generation. Specifically, SPARK introduces a lightweight participation policy that dynamically modulates the contribution of FFN-mediated parametric knowledge while keeping the backbone LLM frozen. The participation policy is optimized with a faithfulness-oriented reward, enabling fine-grained control over parametric knowledge participation during generation. Extensive experiments on six datasets validate the effectiveness of SPARK, which introduces and optimizes only 0.01\% additional trainable parameters while keeping the backbone LLM fully frozen, thereby preserving its general capabilities and achieving more faithful generation than competitive baselines. All datasets and code are available at https://github.com/OpenBMB/SPARK.

cs.CL · cs.AI

新论文 · v1arXiv:2611.26156

Towards Scale-Up Networking: Challenges and Open Questions

Tingting Xu, Zixun Su, Xinyang Huang, Bowen Liu, Kun Tan, Dong Lin, Hao Wang, Yinben Xia 等 12 位作者

原文摘要

The rapid growth of AI models and the widespread use of distributed training and inference have made inter-accelerator communication a critical system bottleneck. Scale-up fabrics address this bottleneck by combining high per-accelerator bandwidth with low and predictable latency over tightly coupled accelerator domains. However, the appropriate topology, protocol mechanisms, communication semantics, and system integration remain actively debated. We analyze the critical-path communication generated by parallelism strategies and derive the resulting requirements for domain scale, per-accelerator bandwidth, and collective completion latency. We then use these requirements as a framework to examine representative proprietary and open scale-up fabrics, revealing how their physical interconnects, topologies, protocols, and semantics make different trade-offs among performance, scalability, compatibility, and implementation cost. Moreover, based on this analysis and discussions with industry experts, we identify challenges and open problems in physical interconnects and switching, efficient communication, semantics, resource sharing and heterogeneous device interconnection. This paper provides a requirement-driven foundation for analyzing current scale-up fabrics and guiding the design of future systems.

cs.NI · cs.DC

版本更新 · v3arXiv:2607.00978

Depth-Only Open-Vocabulary 3D Semantic Segmentation For Privacy-Preserving Robotic Applications

Xuying Huang, Sicong Pan, Maren Bennewitz

原文摘要

Privacy-preserving perception is increasingly important for robotic systems operating in real-world indoor environments, yet it remains underexplored in open-vocabulary 3D semantic segmentation. We study this problem under an RGB-prohibited deployment setting motivated by scene-specific visual information disclosure, where real RGB observations are unavailable during scene acquisition and fusion. To reflect this deployment constraint on existing 3D datasets, we adopt a stricter depth-only evaluation protocol that re-runs scene fusion without RGB and exposes only the resulting depth-derived geometry to the segmentation pipeline. This constraint removes appearance cues that are often critical for open-vocabulary recognition, making depth-only predictions more uncertain and less reliable. To address this challenge, we propose UTTO, a model-agnostic uncertainty-guided test-time optimization framework that uses structured predictive uncertainty as a reliability signal to refine predictions from frozen open-vocabulary 3D backbones. Experiments across ScanNet and Matterport3D demonstrate consistent improvements over multiple depth-only backbones. Privacy recoverability analyses and a real-robot semantic goal grounding case study further support the proposed privacy-constrained setting and applicability.

cs.CV · cs.RO

新论文 · v1arXiv:2611.26107

HiUni: Harmonizing Understanding and Generation in Shared Visual Space

Jiahao Hu, Jingdong Du, Yichen Zhang, Yi Song, Zhenyu Zhang, Jiaxin Niu, Ruihang Li, Xiangzhao Hao 等 13 位作者

原文摘要

Unified multimodal models aim to support visual understanding and generation within a single framework. However, prevailing approaches often separate their visual representations or backbone computation to accommodate the different requirements of these two tasks, which limits the degree of unification. In this paper, we present HiUni, which revisits this design choice and explores whether a pretrained vision language model can be directly extended into a unified model with extensive parameter sharing. Specifically, HiUni reuses the same language-aligned ViT and dense Transformer backbone for both capabilities, while confining task-specific computation to the output side. Yet such a design introduces two key challenges. Generation in pixel space is difficult to optimize, and semantic abstraction in pretrained ViT discards fine spatial details required for synthesis. To address these challenges, HiUni introduces a \emph{Pixel Diffusion Head} tailored to direct pixel-space optimization and a lightweight \emph{HiGate} pathway that provides fine-grained spatial details near the output stage. Extensive experiments show that HiUni achieves a similar average understanding score to the pretrained VLM while attaining competitive performance on image generation and editing. Controlled comparisons and analyses further demonstrate that a single dense backbone can effectively support both capabilities without task-specific backbone computation, highlighting a simpler path toward unified multimodal modeling.

cs.CV · cs.AI

版本更新 · v2arXiv:2611.15073

When Does Causality Help in Multivariate Time-Series Anomaly Detection?

Long Zhang, Wei Huang, Aishui Rao, Jian Hu, Binxiang Wu, Jie Li, Chen Gong, Linxi Feng

原文摘要

Causality-aware scoring is common in multivariate time-series anomaly detection, yet the literature rarely states when it helps. Holding one causality-guided autoencoder (DeCauAE) fixed, we vary graph quality, anomaly type and protocol against ground truth. Graph quality fixes the payoff: on a twenty-variable DAG the gain over a graph-free autoencoder decays as the graph is corrupted. Anomaly type fixes the target: the structural factor lifts window F1 from 0.800 to 0.976 where anomalies break conditional structure, and its advantage vanishes below a break severity s 0.3. Protocol fixes the ranking: point-adjusted F1 lets a random score match the full model, while range-aware VUS separates the autoencoder family from the classical baselines. On real benchmarks the causal module pays only where density makes a system eligible and cross-subsample reproducibility selects entities inside it; enabling it everywhere costs 0.012 VUS-ROC. The structural factor also transfers to classical detectors .

cs.LG

新论文 · v1arXiv:2611.26068

Physics-Informed Deep Learning for Channel Prediction in Satellite-Mounted Metasurface Reflection Communication Networks

Aolin Sun, Ruihong Jiang, Jincong Mo, Yangbo Guo, Zhuangzhuang Cui

原文摘要

Satellite-mounted metasurfaces (MS) provide a low-power programmable paradigm for 6G satellite-terrestrial networks, effectively compensating for severe free-space path loss under stringent payload constraints. However, accurate channel prediction for such systems remains fundamentally challenging due to the coupling of dynamic orbital topology, stochastic user geometry, and hardware-imposed phase quantization. This paper proposes a physics-informed deep learning (PIDL) framework that embeds the physics of wireless communication, including monotonicity constraints and asymptotic decay properties, into a deep neural network for outage probability (OP)-aware channel prediction. We construct a realistic system model incorporating spherical user distribution, Spherical Poisson Point Process-based satellite constellation, and practical metasurface quantization. Simulation results demonstrate that the proposed framework achieves over 0.99 prediction accuracy with physically consistent outage trends under varying geometric configurations, constellation densities, and quantization resolutions.

cs.IT

新论文 · v1arXiv:2611.26055

When Do Rankings Beat Signs? Gradient Direction Recovery under Even Nuisance

Zhuangzhuang Pan, Yan Xia, Chee Seng Chan

原文摘要

Can ranking a batch of antipodal query pairs recover a gradient direction with fewer evaluations than using within-pair signs alone? We study fixed-center recovery in $d$ dimensions for a linear signal plus an unknown differentiable even nuisance, counting function evaluations and feedback rounds separately. Shared pair offsets create an information hierarchy: Mirror retains within-pair signs, Reverse couples reversed cross-pair orders to cancel offsets, and Full enforces all ranking inequalities. Relative shell variation is the nuisance range over the query sphere normalized by the linear-signal scale. When it is at most $c\varepsilon/\sqrt{d}$, Reverse localizes every consistent direction to angular accuracy $\varepsilon$ using $\widetilde{O}(d+\sqrt{d/\varepsilon})$ evaluations in one round at fixed confidence, without a variation bound supplied to the decoder. This matches a lower bound up to logarithmic factors, versus the $Θ(d/\varepsilon)$ one-round sign-only minimax cost on the same class. Over well-conditioned positive-definite quadratics with no common relative-curvature bound, the $R$-round minimax cost is instead $Θ(dR(\varepsilon^{-1/R}-1))$, even with cumulative rankings. We extend validity beyond exact evenness under supplied error bounds. Matched-endpoint experiments certify information beyond signs and short cycles. On two ridge tasks under joint evaluation and round caps, short-cycle recovery attains lower terminal mean gaps than the tested batch-capable controls. An additional optimization benefit from fitting the complete ranking remains unestablished.

cs.LG

新论文 · v1arXiv:2611.26040

Streaming-AAA: Streaming AAA Game Video Generation at High Resolution, High Frame Rate, and Low Latency

Zhihao Hu, Jianqi Ma, Jun Zhang, Jiayang Bai, Haifeng Shi, Yanxiang Zhuang, Zhongqian Sun, Weidong Zhang 等 9 位作者

原文摘要

Controllable streaming video generation offers a promising route toward interactive game simulation, yet existing methods remain far from practical gameplay. They often fail to jointly achieve high visual fidelity, high frame rates, and low interaction latency, resulting in poor responsiveness and degraded user experience. We present Streaming-AAA, a new paradigm for controllable infinite generation of AAA-quality game videos. Our system generates 1080p gameplay videos at 43 FPS with 198 ms end-to-end latency interaction on merely NVIDIA A800 GPUs, while supporting long-horizon play without severe temporal degradation or collapse. Streaming-AAA is built upon three key designs. First, we redesign the VAE architecture for efficient high-resolution decoding, enabling real-time synthesis at 1080p. Second, we pretrain the diffusion model using diffusion forcing and significantly extend the temporal training horizon, allowing the model to capture long-term dependencies and maintain persistent memory. Third, we perform post-training distillation and memory compression to improve inference efficiency while retaining visual quality and temporal consistency. Experiments show that Streaming-AAA substantially improves visual fidelity, frame rate, latency, and long-term stability over prior controllable video generation approaches, enabling a significantly more playable experience for interactive AAA game video generation.

cs.CV

新论文 · v1arXiv:2611.25989

Policy Improvement-Guided Replay for Continual Reinforcement Learning

Qin-Wen Luo, Haoxuan Gao, Sheng Ren, Xiang Chen, Sheng-Jun Huang

原文摘要

Continual reinforcement learning (CRL) requires models to acquire new capabilities while retaining performance on previous tasks. A stability-oriented perspective common in conventional continual learning emphasizes knowledge preservation and interference mitigation. Complementing this perspective, we revisit CRL for vision-language models through policy improvement, a fundamental principle of RL, to preserve acquired performance gains. Using the performance difference lemma, we characterize historical performance changes through the current policy's expected advantages on historical task distributions, motivating replay of historical prompts with the current policy. To allocate limited replay memory, we propose Contrastive Likelihood Replay (CoLiR), which contrasts likelihood gains on correct responses from the final task policy with likelihood reductions on incorrect responses from the initial policy. Our analysis links correct-response likelihood gains to reference-policy advantages under an idealized KL-regularized optimal update and provides an endogenous-reward interpretation for general policy updates. CoLiR estimates scores from finite training rollouts and cached log-likelihoods, selects prompts at task boundaries, and replays them uniformly during subsequent training, with KL regularization restricted to replayed prompts. Memory construction only requires a small amount of additional cross-policy scoring but no additional response generation. Across three Qwen3-VL model sizes on MRCL, CoLiR achieves the highest mean final accuracy and mean task accuracy among the compared methods while maintaining competitive new-task performance.

cs.LG · cs.AI · cs.CL

新论文 · v1arXiv:2611.25986

Seeing Beyond the Obvious: Complementary Semantic Guidance for Salient Single-Positive Multi-Label Learning

Tong Jin, Meng-Long Wang, Xingyu Zhao, Sheng-Jun Huang

原文摘要

Salient Single-Positive Multi-Label Learning (SalSPML) assigns each image its most salient positive label, leaving other classes unlabeled. Existing methods construct class-level Salient Prototypes from observed positives to guide the recognition of unobserved classes. However, since salient annotations favor visually prominent instances, these prototypes may not fully capture the visual diversity of hidden positives, resulting in insufficient positive supervision. To address this limitation, we propose CoST, a Coverage-guided Complementary Soft-Target Refinement framework that supplements Salient Prototypes with complementary class-semantic information. Specifically, CoST identifies semantically relevant candidates from unknown classes by leveraging task-learned category information and image-text semantic correspondence. The selected candidates are then aggregated into class-wise Coverage Prototypes to complement the semantic representations provided by Salient Prototypes. Furthermore, we introduce a ranking-based selective refinement strategy that identifies classes receiving stronger semantic support from Coverage Prototypes and selectively enhances their soft targets while preserving the base soft targets for the rest. Experiments on multiple benchmarks demonstrate the effectiveness of CoST, with strong performance gains over existing baselines.

cs.LG

新论文 · v1arXiv:2611.25981

Safe Subgame Resolving in Multiplayer Games

Boning Li, Longbo Huang

原文摘要

Poker and other games with hidden information are too large to solve in one pass. Strong systems compute a complete blueprint strategy offline and refine it in real time inside the reached subgame. In 2p0s games, a safety theorem guarantees that subgame resolving never increases how much an adversary can gain by exploiting the resolved strategy relative to the blueprint strategy. With three or more players acting independently, no such guarantee exists yet, and heuristic resolving can increase the profile's best-response gap. We prove the first conditional safety theorem for uncorrelated multiplayer subgame resolving, a post-hoc certificate for any delivered update. The best-response gap of the resolved profile exceeds that of the blueprint by at most four terms. Two concern the resolver: the subgame value the blueprint promised without realizing ($A_i^{S}$) and the resolver's subgame solve error ($\varepsilon_i$). The other two are the opponents' best-response gains inside the subgame over their promised values ($\sum_q M_{q}$) and a structural tax ($Θ$), which measures how much the players' total utility can change. The tax is zero in constant-sum games, including every rake-free poker variant. Constructions attain each term with equality. On three-player Leduc poker and on Kuhn poker with up to six players, resolving without a certificate raises the gap beyond a small tolerance in 12% to 100% of the updates. The certificate flags every such update and accepts 41% (Kuhn) and 86% (Leduc) of the standard resolving updates. A resolver that keeps the blueprint whenever the certificate exceeds the tolerance delivers no harmful update. The certificate needs only one best-response pass per opponent inside the subgame, so real-time multiplayer solvers that refine a blueprint can attach it as a safety check. Code is available at https://github.com/lbn187/MultiPlayerSolver

cs.GT · cs.AI

新论文 · v1arXiv:2611.25963

MorphTwin: Enriching Deformable Material Responses from Videos for Morphable and Interactive Physical Twins

Haoqin Hong, Ding Fan, Fubin Dou, Tianyu Huang, Yuzhuo Chen, Zhemeng Zhang, Hanxiao Jiang, Congcong Zhu 等 11 位作者

原文摘要

Reconstructing interactive physical twins from video requires accurately capturing how objects deform while maintaining efficient simulation. We observe that reproducing motion in a video does not guarantee correct responses to new forces when the physical representation restricts how an object can deform. In spring-based twins, coupling between volume and shape responses creates such a limitation. We present MorphTwin, which enriches local deformation modes to capture and independently control these responses while retaining standard spring forces and explicit simulation. To support interpretable material editing and transfer, we use offline calibration to map shared relative response changes to stiffness updates on each reconstructed graph. Under identical inverse settings on synthetic scenes, MorphTwin reduces the mean absolute error of Poisson's ratio by 75.8% and the mean relative error of Young's modulus by 22.1% compared to PhysTwin. We introduce an egocentric stereo and RGB-D benchmark covering interactions with objects of varying stiffness. Real-world experiments demonstrate improved reconstruction accuracy and up to 14.4% lower future tracking error than PhysTwin at interactive speeds.

cs.CV