arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Chinese Academy of Sciences(中国科学院大学)

共收录 1960
2508.15214 2026-02-11 cs.CL

Self-Guided Function Calling in Large Language Models via Stepwise Experience Recall

通过逐步经验回忆实现大语言模型的自引导函数调用

Sijia Cui, Aiyao He, Shuai Xu, Hongming Zhang, Yanna Wang, Qingyang Zhang, Yajing Wang, Bo Xu

机构 * The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences(认知与决策智能复杂系统重点实验室,自动化研究所,中国科学院) School of Artificial Intelligence, University of Chinese Academy of Sciences(人工智能学院,中国科学院大学) Nanjing University of Information Science & Technology(南京信息工程大学) Institute of Computing Technology, Chinese Academy of Sciences(计算技术研究所,中国科学院)

AI总结 SEER通过逐步经验回忆方法,提升大语言模型在多步骤工具使用中的准确性和效率。

Comments Accepted to EMNLP 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.13027 2026-02-11 cs.LG

Deconstructing Positional Information: From Attention Logits to Training Biases

解构位置信息:从注意力logits到训练偏见

Zihan Gu, Ruoyu Chen, Han Zhang, Hua Zhang, Yue Hu

机构 * Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所) School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院) Antai College of Economics and Management, Shanghai Jiao Tong University(上海交通大学安泰经济管理学院)

AI总结 本文通过分析注意力logits和位置编码形式,揭示了乘法编码在整合位置与语义信息上的优势,并发现训练过程中存在的单头沉积模式。

Comments Accepted by ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.08828 2026-02-10 cs.CV

VideoVeritas: AI-Generated Video Detection via Perception Pretext Reinforcement Learning

VideoVeritas:通过感知预设强化学习实现AI生成视频检测

Hao Tan, Jun Lan, Senyuan Shi, Zichang Tan, Zijian Yu, Huijia Zhu, Weiqiang Wang, Jun Wan, Zhen Lei

机构 * MAIS, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所MAIS部) School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院) Shenzhen Institute of Advanced Technology (SIAT), Chinese Academy of Sciences(中国科学院深圳先进技术研究所)

AI总结 VideoVeritas通过感知预设强化学习提升AI生成视频检测性能,整合细粒度感知与事实推理,采用时空定位和自监督计数提升检测效果。

Comments Project: https://github.com/EricTan7/VideoVeritas

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.08253 2026-02-10 cs.AI

G-LNS: Generative Large Neighborhood Search for LLM-Based Automatic Heuristic Design

G-LNS:基于生成式大邻域搜索的LLM自动启发式设计

Baoyun Zhao, He Wang, Liang Zeng

机构 * Software College, Northeastern University(东北大学软件学院) International Centre for Theoretical Physics Asia-Pacific, University of Chinese Academy of Sciences(中国科学院大学国际理论物理亚太中心) Taiji Laboratory for Gravitational Wave Universe, University of Chinese Academy of Sciences(中国科学院大学太极引力波宇宙实验室) Tsinghua University(清华大学)

AI总结 G-LNS通过生成式进化框架,结合LLM共同进化破坏与修复操作符,实现了大邻域搜索操作符的自动化设计,在复杂组合优化问题中表现出更强的性能和泛化能力。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.05787 2026-02-10 cs.IR cs.AI cs.CL

Bagging-Based Model Merging for Robust General Text Embeddings

基于Bagging的模型融合用于鲁棒通用文本嵌入

Hengran Zhang, Keping Bi, Jiafeng Guo, Jiaming Zhang, Wenbo Yang, Daiting Shi, Xueqi Cheng

机构 * State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences(人工智能安全国家重点实验室) University of Chinese Academy of Sciences(中国科学院大学) Querit Private Limited(Querit私人有限公司)

AI总结 BOOM通过批量级洗牌和模型融合提升文本嵌入鲁棒性,减少增量学习成本。

Comments 12 pages, 4 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.22896 2026-02-10 cs.AI

Game-Theoretic Co-Evolution for LLM-Based Heuristic Discovery

基于博弈论的LLM启发式发现共演化

Xinyi Ke, Kai Li, Junliang Xing, Yifan Zhang, Jian Cheng

机构 * C2DL, Institute of Automation, Chinese Academy of Sciences(C2DL,自动化研究所,中国科学院) School of Artificial Intelligence, University of Chinese Academy of Sciences(人工智能学院,中国科学院大学) Tsinghua University(清华大学)

AI总结 本文提出ASRO框架,通过博弈论方法实现求解器与实例生成器的共演化,提升启发式发现的泛化能力和鲁棒性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.04093 2026-02-10 cs.CL

SearchAttack: Red-Teaming LLMs against Knowledge-to-Action Threats under Online Web Search

SearchAttack: 对知识到行动威胁下对LLM进行红队攻击

Yu Yan, Sheng Sun, Mingfeng Li, Zheming Yang, Chiwei Zhu, Fei Ma, Benfeng Xu, Min Liu, Qi Li

机构 * Institute of Computing Technology, CAS(计算技术研究所,中国科学院) University of Chinese Academy of Sciences(中国科学院大学) People's Public Security University of China(中国人民公安大学) University of Science and Technology of China(中国科学技术大学) Guangdong Laboratory of Artificial Intelligence and Digital Economy(广东人工智能与数字经济实验室) Tsinghua University(清华大学)

AI总结 SearchAttack通过重新表述有害语义和测试奖励追逐偏差,揭示LLM在搜索增强任务中的安全漏洞。

Comments Misusing LLM-driven search for harmful information-seeking poses serious risks. We characterize its usability and impact through a comprehensive red-teaming and evaluation

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.22522 2026-02-10 cs.LG cs.AI

Towards Reliable Evaluation of Adversarial Robustness for Spiking Neural Networks

面向脉冲神经网络对抗鲁棒性可靠评估的框架

Jihang Wang, Dongcheng Zhao, Ruolin Chen, Qian Zhang, Yi Zeng

机构 * Brain-inspired Cognitive AI Lab, Institute of Automation, Chinese Academy of Sciences(脑启发认知人工智能实验室,自动化研究所,中国科学院) School of Artificial Intelligence, University of Chinese Academy of Sciences(人工智能学院,中国科学院大学) Beijing Institute of AI Safety and Governance(北京人工智能安全与治理研究院) Beijing Key Laboratory of Safe Al and Superalignment(北京安全人工智能与超对齐重点实验室) Center for Long-term AI(长期人工智能中心)

AI总结 本文提出ASSG框架和SA-PGD攻击方法,提升SNN对抗鲁棒性评估的可靠性与准确性

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.13786 2026-02-10 cs.SD cs.AI

DegDiT: Controllable Audio Generation with Dynamic Event Graph Guided Diffusion Transformer

DegDiT:基于动态事件图引导的扩散变换器的可控音频生成

Yisu Liu, Chenxing Li, Wanqian Zhang, Wenfu Wang, Meng Yu, Ruibo Fu, Zheng Lin, Weiping Wang, Dong Yu

机构 * Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所) School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院) Tencent, AI Lab(腾讯AI实验室) Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所)

AI总结 DegDiT通过动态事件图引导的扩散变换器框架,实现了开放式词汇可控音频生成,提升了生成音频的准确性和多样性。

Journal ref IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.07804 2026-02-10 cs.CL cs.AI

Pruning as a Cooperative Game: Surrogate-Assisted Layer Contribution Estimation for Large Language Models

剪枝作为合作博弈:用于大型语言模型的代理辅助层贡献估计

Xuan Ding, Pengyu Tong, Ranjie Duan, Yunjian Zhang, Rui Sun, Yao Zhu

机构 * Shenzhen Future Network of Intelligence Institute(深圳未来网络智能研究院) Guangdong Provincial Key Laboratory of Future Networks of Intelligence(广东省未来网络智能重点实验室) The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳)) Beijing Normal University(北京师范大学) Alibaba Group(阿里巴巴集团) University of Chinese Academy of Sciences(中国科学院大学) Zhejiang University(浙江大学)

AI总结 本文提出基于博弈论的层剪枝方法,利用代理网络估计层贡献,实现高效且有效的大型语言模型剪枝。

Comments Accepted by ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.07498 2026-02-10 cs.CV

IM-Animation: An Implicit Motion Representation for Identity-decoupled Character Animation

IM-Animation: 一种隐式运动表示用于身份解耦的角色动画

Zhufeng Xu, Xuan Gao, Feng-Lin Liu, Haoxian Zhang, Zhixue Fang, Yu-Kun Lai, Xiaoqiang Liu, Pengfei Wan, Lin Gao

机构 * Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所) University of Chinese Academy of Sciences(中国科学院大学) Kling Team, Kuaishou Technology(快手技术 Kling 团队) Cardiff University(卡迪夫大学)

AI总结 IM-Animation通过隐式运动表示和时间一致的遮罩令牌重定向模块,实现身份解耦的角色动画生成,提升动画质量和训练效率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.16010 2026-02-10 cs.SD cs.AI eess.AS

Fed-PISA: Federated Voice Cloning via Personalized Identity-Style Adaptation

Fed-PISA: 通过个性化身份-风格适应实现联邦语音克隆

Qi Wang, Shituo Ma, Guoxin Yu, Hanyang Peng, Yue Yu

机构 * Peng Cheng Laboratory(鹏城实验室) Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所) Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所) University of Chinese Academy of Sciences(中国科学院大学)

AI总结 Fed-PISA通过个性化身份-风格适应实现联邦语音克隆,减少通信成本并提升个性化表现。

Comments Accepted by ICASSP 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.21001 2026-02-10 cs.LG cs.AI

LEASE: Offline Preference-based Reinforcement Learning with High Sample Efficiency

LEASE: 基于偏好的离线强化学习与高样本效率

Xiao-Yin Liu, Guotao Li, Xiao-Hu Zhou, Zeng-Guang Hou

机构 * State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences(多模态人工智能系统国家重点实验室,中国科学院自动化研究所) School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院) CASIA-MUST Joint Laboratory of Intelligence Science and Technology, Institute of Systems Engineering, Macau University of Science and Technology(智能科学与技术CASIA-MUST联合实验室,澳门科学理工学院)

AI总结 LEASE通过高样本效率的离线偏好强化学习算法,在减少偏好数据需求的同时提升奖励模型的准确性与稳定性。

Comments Accepted by IEEE Transactions on Systems, Man, and Cybernetics: Systems

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.07060 2026-02-10 eess.IV cs.CV physics.ins-det physics.med-ph

U-Net Based Image Enhancement for Short-time Muon Scattering Tomography

基于U-Net的短时μ子散射断层成像图像增强

Haochen Wang, Pei Yu, Liangwen Chen, Weibo He, Yu Zhang, Yuhong Yu, Xueheng Zhang, Lei Yang, Zhiyu Sun

机构 * School of Physics, Hefei University of Technology(合肥工业大学物理学院) Advanced Energy Science and Technology Guangdong Laboratory(广东实验室) Institute of Modern Physics, Chinese Academy of Sciences(中国科学院现代物理研究所) School of Nuclear Science and Technology, University of Chinese Academy of Sciences(中国科学院大学核科学技术学院) State Key Laboratory of Heavy Ion Science and Technology, Institute of Modern Physics, Chinese Academy of Sciences(中国科学院现代物理研究所重离子科学与技术国家重点实验室) Institute of Materials, China Academy of Engineering Physics(工程物理院材料研究所)

AI总结 本文提出基于U-Net的图像增强方法,利用模拟数据训练提升短时μ子散射断层成像的图像质量,实验显示SSIM显著提高,LPIPS显著降低,为实际应用奠定基础。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.06968 2026-02-10 cs.RO cs.CV

Learning to Anchor Visual Odometry: KAN-Based Pose Regression for Planetary Landing

学习锚定视觉里程计:基于KAN的行星着陆姿态回归

Xubo Luo, Zhaojin Li, Xue Wan, Wei Zhang, Leizheng Shu

机构 * University of Chinese Academy of Sciences(中国科学院大学) Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences(中国科学院空间利用技术与工程中心)

AI总结 KANLoc通过基于KAN的姿态回归器和混合定位方案,实现高精度、实时的行星着陆定位,有效减少漂移并提升鲁棒性。

Comments 8 pages, accepted by RA-L

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.06529 2026-02-09 cs.CV

AdaptOVCD: Training-Free Open-Vocabulary Remote Sensing Change Detection via Adaptive Information Fusion

AdaptOVCD: 无需训练的开放词汇遥感变化检测 via 自适应信息融合

Mingyu Dou, Shi Qiu, Ming Hu, Yifan Chen, Huping Ye, Xiaohan Liao, Zhe Sun

机构 * Key Laboratory of Spectral Imaging Technology CAS, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences (CAS)(光谱成像技术重点实验室、光学与精密机械研究所、中国科学院) University of Chinese Academy of Sciences(中国科学院大学) Institute of Artificial Intelligence (TeleAI), China Telecom(人工智能研究所(TeleAI),中国电信) College of Future Information Technology, Fudan University(复旦大学未来信息技术学院) State Key Laboratory of Resources and Environment Information System, Institute of Geographic Sciences and Natural Resources Research, CAS(资源与环境信息系统国家重点实验室、地理科学与自然资源研究所、中国科学院) Key Laboratory of Low Altitude Geographic Information and Air Route, CAAC(低空地理信息与航线重点实验室、民航局) School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University(人工智能、光学与电子学院(iOPEN),西北工业大学)

AI总结 AdaptOVCD通过自适应信息融合实现无需训练的开放词汇遥感变化检测,有效缓解误差传播并提升检测性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.06380 2026-02-09 cs.RO cs.SY eess.SY

A Consistency-Improved LiDAR-Inertial Bundle Adjustment

改进一致性 的 LiDAR-惯性束调整

Xinran Li, Shuaikang Zheng, Pengcheng Zheng, Xinyang Wang, Jiacheng Li, Zhitian Li, Xudong Zou

机构 * Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院航空信息研究所) University of Chinese Academy of Sciences(中国科学院大学) Aerospace Information Technology University(航空信息科技大学) QiLu Aerospace Information Research Institute, Chinese Academy of Sciences(齐鲁航空信息研究所) School of Electronic, Electrical and Communication Engineering(电子电气与通信工程学院)

AI总结 本文提出改进一致性的 LiDAR-惯性束调整方法,通过定制参数化和估计器提升 SLAM 系统的准确性和可观测性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.05444 2026-02-09 cs.CL

Causal Front-Door Adjustment for Robust Jailbreak Attacks on LLMs

因果前门调整用于对抗大语言模型的鲁棒性劫持攻击

Yao Zhou, Zeen Song, Wenwen Qiang, Fengge Wu, Shuyi Zhou, Changwen Zheng, Hui Xiong

机构 * Institute of Software Chinese Academy of Sciences, Beijing, China(中国科学院软件研究所) University of Chinese Academy of Sciences, Beijing, China(中国科学院大学) Institute of Information Engineering Chinese Academy of Sciences, Beijing, China(中国科学院信息工程研究所) Hong Kong University of Science and Technology, China, Hong Kong, China(香港科学与技术大学)

AI总结 本文提出CFA²攻击方法,通过因果前门准则和稀疏自编码器实现对大语言模型的鲁棒性劫持,提升攻击成功率并提供机制解释。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.04928 2026-02-09 cs.LG

Euphonium: Steering Video Flow Matching via Process Reward Gradient Guided Stochastic Dynamics

euphonium:通过过程奖励梯度引导的随机动态控制视频流匹配

Ruizhe Zhong, Jiesong Lian, Xiaoyue Mi, Zixiang Zhou, Yuan Zhou, Qinglin Lu, Junchi Yan

机构 * Huazhong University of Science and Technology(华中科技大学) Shanghai Jiao Tong University(上海交通大学) Tencent Hunyuan(腾讯混元) University of Chinese Academy of Sciences(中国科学院大学)

AI总结 Euphonium通过过程奖励梯度引导动态,提升视频流匹配的生成效率和对齐性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.02614 2026-02-09 cs.SE cs.AI cs.CR

Testing Storage-System Correctness: Challenges, Fuzzing Limitations, and AI-Augmented Opportunities

存储系统正确性测试:挑战、模糊测试局限性与AI增强的机会

Ying Wang, Jiahui Chen, Dejun Jiang

机构 * Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS)(中国科学院计算技术研究所) University of Chinese Academy of Sciences(中国科学院大学) Zhongguancun Laboratory(中关村实验室)

AI总结 本文探讨了存储系统正确性测试的挑战,分析了模糊测试的局限性,并提出了利用人工智能技术提升测试效果的可能性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2407.14069 2026-02-09 cs.CV

Self-Supervised Video Representation Learning in a Heuristic Decoupled Perspective

基于启发式解耦视角的自监督视频表示学习

Zeen Song, Wenwen Qiang, Changwen Zheng, Hui Xiong, Gang Hua

机构 * National Key Laboratory of Space Integrated Information System, Institute of Software Chinese Academy of Sciences(空间信息集成国家重点实验室,软件研究所中国科学院) University of Chinese Academy of Sciences(中国科学院大学) Hong Kong University of Science and Technology(香港理工大学) Dolby Laboratories Inc(杜比实验室有限公司) Xi’an Jiaotong University(西安交通大学)

AI总结 本文提出BOD-VCL方法,通过解耦静态和动态语义,提升视频对比学习的性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2405.16895 2026-02-09 cs.CV

Anonymization Prompt Learning for Facial Privacy-Preserving Text-to-Image Generation

面向面部隐私保护的匿名化提示学习

Liang Shi, Jie Zhang, Shiguang Shan

机构 * University of Chinese Academy of Sciences(中国科学院大学)

AI总结 本文提出匿名化提示学习方法,通过训练可学习的提示前缀,实现文本到图像生成中面部身份的匿名化,以保护隐私并防止深度伪造风险。

Comments Accepted by IJCV

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.06087 2026-02-09 cs.RO cs.SY eess.SY

Dynamic Modeling, Parameter Identification and Numerical Analysis of Flexible Cables in Flexibly Connected Dual-AUV Systems

柔性双水下机器人系统中柔性缆线的动态建模、参数识别与数值分析

Kuo Chen, Minghao Dou, Qianqi Liu, Yang An, Kai Ren, Zeming WU, Yu Tian, Jie Sun, Xinping Wang, Zhier Chen, Jiancheng Yu

机构 * the State Key Laboratory of Robotics Intelligent Systems, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016 University of Chinese Academy of Sciences, Beijing 100049, China Road Joint Laboratory on Space \& Sea Technology Advanced Research, Shanghai 201304, China Institute of Deep-Sea Science Engineering, Chinese Academy of Sciences, Sanya 572000, China the Innovation Academy for Microsatellites, Chinese Academy of Sciences, Shanghai 201109 University of Nottingham Ningbo China, Ningbo 315199

AI总结 本研究提出了一种用于柔性双水下机器人系统动态建模和参数识别的方法,揭示了柔性缆线在复杂边界条件下的非线性动力学特性,为系统设计和控制提供了理论支持。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.05998 2026-02-06 cs.CV

VisRefiner: Learning from Visual Differences for Screenshot-to-Code Generation

VisRefiner: 从视觉差异中学习以实现截图到代码生成

Jie Deng, Kaichun Yao, Libo Zhang

机构 * Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所) University of Chinese Academy of Sciences(中国科学院大学)

AI总结 VisRefiner通过学习视觉差异提升截图到代码生成的准确性和自我完善能力。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.05577 2026-02-06 cs.CV

LocateEdit-Bench: A Benchmark for Instruction-Based Editing Localization

LocateEdit-Bench:基于指令的编辑本地化基准测试

Shiyu Wu, Shuyan Li, Jing Li, Jing Liu, Yequan Wang

机构 * Institute of Automation, Chinese Academy of Sciences, Beijing, China(中国科学院自动化研究所) Beijing Academy of Artificial Intelligence, Beijing, China(北京人工智能研究院) University of Chinese Academy of Sciences, Beijing, China(中国科学院大学) School of EEECS, Queen's University Belfast, Northern Ireland, UK(女王大学 Belfast 电子工程与计算机科学学院) Harbin Institute of Technology, Shenzhen, China(哈尔滨工业大学(深圳)) Peking University, Beijing, China(北京大学)

AI总结 LocateEdit-Bench是一个针对基于指令的编辑本地化问题设计的大规模数据集,整合了四种先进编辑模型,用于评估和推动伪造本地化方法的发展。

Comments 11 pages, 7 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.15176 2026-02-06 cs.CV

SpikeGS: Learning 3D Gaussian Fields from Continuous Spike Stream

SpikeGS: 从连续尖峰流中学习3D高斯场

Jinze Yu, Xin Peng, Zhengda Lu, Laurent Kneip, Yiqun Wang

机构 * College of Computer Science, Chongqing University, Chongqing, China(重庆大学计算机学院) Motovis Co., Ltd., Shanghai, China(Motovis公司) University of Chinese Academy of Sciences, Beijing, China(中国科学院大学) Mobile Perception Lab, ShanghaiTech University, Shanghai, China(上海科技大学移动感知实验室)

AI总结 SpikeGS通过从连续尖峰流中学习3D高斯场,实现了高鲁棒性和高质量的实时渲染,适用于低光环境下精细纹理的视图合成。

Comments Accepted by ACCV 2024

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.04365 2026-02-05 cs.LG

EXaMCaP: Subset Selection with Entropy Gain Maximization for Probing Capability Gains of Large Chart Understanding Training Sets

EXaMCaP: 通过熵增最大化进行子集选择以探测大规模图表理解训练集的探测能力提升

Jiapeng Liu, Liang Li, Bing Li, Peng Fu, Xiyan Gao, Chengyang Fang, Xiaoshuai Hao, Can Ma

机构 * Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China(中国科学院信息工程研究所) School of Cyberspace Security, University of Chinese Academy of Sciences, Beijing, China(中国科学院大学网络安全学院) School of Computer and Artificial Intelligence, Jiangxi University of Finance and Economics(江西财经大学计算机与人工智能学院)

AI总结 EXaMCaP通过熵增最大化选择子集,以高效探测大规模图表理解训练集的能力提升。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.09465 2026-02-05 cs.AI

EvoFSM: Controllable Self-Evolution for Deep Research with Finite State Machines

EvoFSM: 通过有限状态机实现可控的自我进化用于深度研究

Shuo Zhang, Chaofa Yuan, Ryan Guo, Xiaomin Yu, Rui Xu, Zhangquan Chen, Zinuo Li, Zhi Yang, Shuhao Guan, Zhenheng Tang, Sen Hu, Liwen Zhang, Ronghao Chen, Huacan Wang

机构 * QuantaAlpha HKUST(GZ)(香港科技大学(广州)) FDU(福建大学) THU(清华大学) SUFE(上海财经大学) UCD(都柏林大学) HKUST(香港科技大学) PKU(北京大学) UCAS(中国科学技术大学)

AI总结 EvoFSM通过有限状态机实现可控自我进化,提升智能体在开放性问题中的适应性与稳定性,有效解决传统方法的不稳定性与幻觉问题。

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.17774 2026-02-05 cs.CV cs.GR

Quasi-Medial Distance Field (Q-MDF): A Robust Method for Approximating and Discretizing Neural Medial Axes

准中轴距离场(Q-MDF):一种用于近似和离散化神经中轴的稳健方法

Jiayi Kong, Chen Zong, Jun Luo, Shiqing Xin, Fei Hou, Hanqing Jiang, Chen Qian, Ying He

机构 * S-Lab, Nanyang Technological University(南洋理工大学S实验室) School of Mathematics, Nanjing University of Aeronautics and Astronautics(南京航空航天大学数学学院) School of Computer Science and Technology, Shandong University(山东大学计算机科学与技术学院) Key Laboratory of System Software (CAS), Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所系统软件重点实验室) University of Chinese Academy of Sciences(中国科学院大学) SenseTime Research(商汤科技研究院)

AI总结 本文提出Q-MDF方法,通过隐式重建技术实现从点云和网格中稳健提取中轴,提升中轴变换的准确性和鲁棒性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2305.19557 2026-02-05 math.OC cs.LG eess.SP stat.ML

Dictionary Learning under Symmetries via Group Representations

通过群表示学习对称性的字典

Subhroshekhar Ghosh, Aaron Y. R. Low, Yong Sheng Soh, Zhuohang Feng, Brendan K. Y. Tan

机构 * Department of Mathematics, National University of Singapore(新加坡国立大学数学系) School of Mathematical Sciences, University of Chinese Academy of Sciences(中国科学院大学数学科学学院)

AI总结 本文通过群表示理论,提出了一种学习具有对称性不变性的字典的方法,并在SO(2)和SO(3)群上进行了验证。

Comments 33 pages, 3 figures

详情

展开后加载摘要…

URL PDF HTML 收藏