arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

Peking University(北京大学)

共收录 239
2602.01051 2026-07-23 cs.LG 版本更新

SwiftRepertoire: Few-Shot Immune-Signature Synthesis via Dynamic Kernel Codes

SwiftRepertoire: 通过动态核码实现少样本免疫特征合成

Rong Fu, Yang Li, Yabin Jin, Jiekai Wu, Chunlei Meng, Youjin Wang

机构 * University of Macau(澳门大学) Peking University(北京大学) University of Chinese Academy of Sciences(中国科学院大学) The First People’s Hospital of Foshan(佛山第一人民医院) Juntendo University(立命馆大学) Shanghai AI Laboratory(上海人工智能实验室) Fudan University(复旦大学) Renmin University of China(中国人民大学) Minzu University of China(民族大学) Tsinghua University(清华大学) Capital Medical University(首都医科大学)

AI总结 本文提出SwiftRepertoire框架,通过动态核码实现少样本免疫特征合成,利用轻量任务描述符和原型字典生成紧凑任务特定参数化,实现快速适应新任务,提升临床和研究场景下的模型实用性与可解释性。

Comments 19 pages, 8 figures, 8 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.10481 2026-07-22 cs.LG cs.AI cs.CL 版本更新

ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples

ARMOR:使用离策略锚样本稳定在线大语言模型强化学习

Kexin Huang, Junkang Wu, Jinda Lu, Shuo Yang, Chiyu Ma, Jiancan Wu, Xiang Wang, Xiangnan He, Guoyin Wang, Jingren Zhou

机构 * University of Science and Technology of China(中国科学技术大学) Peking University(北京大学) Dartmouth College(达特茅斯学院)

AI总结 研究大语言模型强化学习训练不稳定问题,提出ARMOR框架,通过锚展开利用离策略数据保留解决方案模式,混合优化重新制定策略目标实现可控探索,经实验验证可有效减轻验证崩溃,提升性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.04733 2026-07-22 cs.CL cs.LG 版本更新

LP-SFT: Local-Preserving Supervised Fine-Tuning via Multimodal Entropy Structure

LP-SFT:通过多模态熵结构进行局部保持监督微调

Yueyang Wang, Baolong Bi, Shuo Lu, Jingyuan Zhang, Jiajun Shi

机构 * School of Mathematical Sciences, Peking University(北京大学数学科学学院) Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所) Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所) College of Computing, Georgia Institute of Technology(佐治亚理工学院计算学院)

AI总结 研究监督微调中存在的问题,利用香农和雷尼熵分析揭示预训练模型的多模态熵结构,提出LP-SFT目标,在多领域实验中提升性能,平衡准确率和k准确率。

Comments 20 pages, 3 figures. Code is available at https://github.com/Wakaka161/LP-SFT

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.14469 2026-07-22 cs.CL 版本更新

Measuring and Mitigating Post-hoc Rationalization in Reverse Chain-of-Thought Generation

反向思维链生成中的事后合理化测量与缓解

Guangyue Peng, Zongchao Chen, Wen Luo, Yuntao Wen, Wei Li, Ruixiang Feng, Ran Le, Chen Yang, Zhenwei An, Yang Song, Tao Zhang, Houfeng Wang

机构 * State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University(信息处理国家重点实验室,计算机科学学院,北京大学) University of Electronic Science and Technology of China(电子科技大学) Nanbeige Lab, BOSS Zhipin(纳贝格实验室,BOSS智联)

AI总结 针对反向思维链生成中的事后合理化问题,提出结构骨架引导推理方法,通过解耦答案依赖而非抑制答案来缓解锚定效应,在开放推理基准上提升10%性能。

Comments ICML 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.13597 2026-07-21 cs.RO cs.AI cs.CV 版本更新

Semantic Anchoring for Robotic Action Representations

用于机器人动作表示的语义锚定

Yuan Xu, Youheng Shi, Chengyang Li, Wentao Zhu, Yizhou Wang

机构 * Peking University(北京大学) Eastern Institute of Technology, Ningbo(宁波东方理工大学) Shanghai Jiao Tong University(上海交通大学)

AI总结 研究VLA模型微调后动作表示结构受损问题,受镜像神经元理论启发,通过系统探测证实结构变化与任务表现相关。提出即插即用方法,将动作表示锚定到语义流形并分解通道,经多基准测试验证,有效提升了模型在真实世界任务中的表现。

Comments Project Page: https://xy02-05.github.io/SemanticMN

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.12071 2026-07-21 cs.CL 版本更新

Beyond Parallel Tracking: Interactive Multi-Feature Fusion Drives Semantic Reconstruction from Non-invasive Brain Recordings

超越并行跟踪:交互式多特征融合驱动非侵入性脑记录的语义重建

Boda Xiao, Xiran Xu, Songyi Li, Yujie Yan, Xihong Wu, Heping Cheng, Jing Chen

机构 * Center for BioMed-X Research, Academy for Advanced Interdisciplinary Studies,Peking University(北京大学生物医学前沿创新中心、前沿交叉学科研究院) Speech and Hearing Research Center, School of Intelligence Science and Technology,Peking University(北京大学智能科学与技术学院言语听觉研究中心) State Key Laboratory of General Artificial Intelligence(通用人工智能国家重点实验室) National Biomedical Imaging Center, State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, College of Future Technology, Peking University(国家生物医学成像中心、膜生物学国家重点实验室、分子医学研究所、北京大学生命科学联合中心、未来技术学院)

AI总结 研究针对非侵入性脑记录语义重建中表征不匹配问题,引入多特征融合框架,通过交互式门控机制结合静态与动态表征,经实验对比线性连接和非线性交叉注意力等方法,证明交叉注意力融合性能最佳,提供了新的脑到文本解码方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.04438 2026-07-21 cs.CV cs.AI cs.HC cs.MA cs.MM 版本更新

ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog

ResearchStudio-Reel:实现从论文到海报、视频和博客的研究最后一公里自动化

Lingao Xiao, Yalun Dai, Yangyu Huang, Qihao Zhao, Wenshan Wu, Hugo He, Ruishuo Chen, Jin Jiang, Qianli Ma, Jiahuan Zhang, Xin Zhang, Ying Xin, Yang Ou, Yan Xia, Scarlett Li, Longbo Huang, Zhipeng Zhang, Yang He, Yap Kim Hui, Yan Lu

机构 * Microsoft Research(微软研究院) National University of Singapore(新加坡国立大学) Nanyang Technological University(南洋理工大学) Tsinghua University(清华大学) Peking University(北京大学) Shanghai Jiao Tong University(上海交通大学) Westlake University(西湖大学) CFAR, A*STAR(计算科学与工程研究所,新加坡科技研究局)

AI总结 研究传播自动化困难,以往方法有局限。该研究提出将最后一公里构建为技能组合,实例化ResearchStudio-Reel,包括共享提取器、可编辑生成器和交互式收敛层,能产出多种可编辑工件,效果优于现有系统。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.03043 2026-07-21 cs.CL cs.AI cs.LG 版本更新

Lil: Less is Less When Applying Post-Training Sparse-Attention Algorithms in Long-Decode Stage

Lil: 在长解码阶段应用后训练稀疏注意力算法时,少即是少

Junhao Hu, Fangze Li, Mingtao Xu, Feifan Meng, Shiju Zhao, Tiancheng Hu, Ting Peng, Anmin Liu, Wenrui Huang, Chenxu Liu, Ziyue Hua, Tao Xie

机构 * SCS, Peking University, Beijing, China(北京大学信息科学与技术学院,北京,中国) Key Lab of HCST (PKU), MOE, Beijing, China(高等教育出版社HCST重点实验室(PKU),北京,中国) State Key Laboratory for Novel Software Technology, Nanjing University, China(南京大学新型软件技术国家重点实验室,中国) Tencent, Shenzhen, China(腾讯,深圳,中国) Beijing Tongming Lake Information Technology Application Innovation Center, Beijing, China(北京 Tongming Lake 信息技术应用创新中心,北京,中国)

AI总结 本文研究了在长解码阶段应用稀疏注意力算法时,信息丢失导致序列变长的问题,提出早停算法减少token消耗并降低精度损失。

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.16461 2026-07-21 cs.CV 版本更新

GAP-MLLM: Geometry-Aligned Pre-training for Activating 3D Spatial Perception in Multimodal Large Language Models

GAP-MLLM:几何对齐预训练以激活多模态大语言模型中的3D空间感知

Jiaxin Zhang, Junjun Jiang, Haijie Li, Youyu Chen, Kui Jiang, Dave Zhenyu Chen

机构 * Harbin Institute of Technology(哈尔滨工业大学) School of Electronic and Computer Engineering, Peking University(北京大学电子与计算机工程学院) Huawei(华为)

AI总结 本文提出GAP-MLLM,通过几何对齐预训练激活多模态大语言模型中的3D空间感知,改进了传统方法在3D空间感知上的不足。

Comments Accepted by ECCV 2026. Project page: https://gapmllm.github.io/

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.18436 2026-07-21 cs.CV 版本更新

When Generative Replay Meets Evolving Deepfakes: Dual Confusion-Aware Regularization for Incremental Face Forgery Detection

当生成式重放遇上不断演变的深度伪造:用于增量式人脸伪造检测的双混淆感知正则化

Hao Shen, Jikang Cheng, Renye Yan, Zhongyuan Wang, Wei Peng, Baojin Huang

机构 * Huazhong Agricultural University(华中农业大学) Peking University(北京大学) Wuhan University(武汉大学) Stanford University(斯坦福大学)

AI总结 研究人脸生成技术发展下的增量式深度伪造检测问题,提出双混淆感知正则化策略双CARE,通过引入域感知混淆分数量化域混淆,对重放生成器和检测器进行双调制,有效利用生成式重放改进检测效果。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.07457 2026-07-21 cs.CL cs.AI 版本更新

GRIP: In-Parameter Graph Reasoning through Fine-Tuning Large Language Models

GRIP:通过微调大语言模型进行参数内图推理

Jiarui Feng, Donghong Cai, Yixin Chen, Muhan Zhang

机构 * Washington University in Saint Louis(华盛顿大学圣路易斯分校) Institute for Artificial Intelligence, Peking University(北京大学人工智能研究院) State Key Laboratory of General Artificial Intelligence, BIGAI(通用人工智能国家重点实验室,BIGAI)

AI总结 研究如何让大语言模型适应结构数据,提出GRIP方法,通过微调任务将图关系知识内化到模型参数,存储于轻量级LoRA模块,实验表明该方法在处理大图时优于基线,处理小图时以低推理成本达可比性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.09371 2026-07-21 stat.ME cs.LG 版本更新

Representation-Aware Distributionally Robust Optimization: A Knowledge Transfer Framework

表示感知分布鲁棒优化:一种知识转移框架

Zitao Wang, Nian Si, Molei Liu

机构 * Department of Statistics, Columbia University(哥伦比亚大学统计系) Department of Industrial Engineering and Decision Analytics, Hong Kong University of Science and Technology(香港科技大学工业工程与决策分析系) Department of Biostatistics, Peking University Health Science Center(北京大学北京医科大学生物统计学系) Beijing International Center for Mathematical Research, Peking University(北京大学北京国际数学研究中心)

AI总结 研究提出表示感知分布鲁棒估计(READ)框架,利用外部表示指导鲁棒性几何,增加改变表示坐标扰动的运输成本。在当前目标推断和未来总体部署中研究READ,模拟和应用证明其在多源多任务转移学习中有优势。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.00898 2026-07-20 cs.LG cs.RO 版本更新

Dichotomous Diffusion Policy Optimization

二元扩散策略优化

Ruiming Liang, Yinan Zheng, Kexin Zheng, Tianyi Tan, Jianxiong Li, Liyuan Mao, Zhihao Wang, Guang Chen, Hangjun Ye, Jingjing Liu, Jinqiao Wang, Xianyuan Zhan

机构 * Fundation Model Research Center, Institute of Automation, Chinese Academy of Sciences(基础模型研究中心,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) Institute for AI Industry Research (AIR), Tsinghua University(人工智能产业研究院(AIR),清华大学) The Chinese University of Hong Kong(香港中文大学) Shanghai Jiao Tong University(上海交通大学) Peking University(北京大学) Xiaomi EV(小米电动车)

AI总结 DIPOLE是一种新的RL算法,通过二元策略分解实现稳定可控的扩散策略优化,适用于复杂现实应用。

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.11068 2026-07-20 cs.LG cs.AI 版本更新

MAnchors: Memorization-Based Acceleration of Anchors via Rule Reuse and Transformation

MAnchors: 通过规则重用和转换基于记忆的锚点加速

Haonan Yu, Junhao Liu, Xin Zhang

机构 * School of Computer Science, Peking University, Beijing, China(北京大学计算机科学学院) Key Lab of High Confidence Software Technologies(Peking University), Ministry of Education, Beijing, China(高可信软件技术重点实验室(北京大学))

AI总结 MAnchors通过规则重用和转换基于记忆的方法加速锚点,减少解释生成时间并保持保真度和可解释性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2303.01421 2026-07-20 cs.CL cs.LG 版本更新

Learn to Memorize: Scalable Continual Learning in Semiparametric Models with Mixture-of-Neighbors Induction Memory

学习记忆:基于邻居混合归纳记忆的半参数模型中的可扩展持续学习

Guangyue Peng, Tao Ge, Wen Luo, Wei Li, Houfeng Wang

机构 * State Key Laboratory for Multimedia Information Processing, School of Computer Science, Peking University(多媒体信息处理国家重点实验室,计算机科学学院,北京大学) Microsoft(微软公司) Microsoft Research Asia(微软亚洲研究院)

AI总结 研究半参数语言模型中记忆缺乏学习能力的问题,提出将非参数记忆重新概念化为可学习的邻居混合归纳记忆(MoNIM),融入模型信息流,经实验验证其能提升半参数语言模型的可扩展性和持续学习性能。

Comments 15 pages, 5 figures

Journal ref Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 28517-28531, Vienna, Austria. Association for Computational Linguistics, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.22433 2026-07-17 cs.LG 版本更新

From physical surfaces to human-centric heat stress: LST and UTCI heat mapping reveals nonlinear effects of urban morphology

超越地表温度:可解释的空间机器学习揭示城市形态对以人类为中心的热压力的影响

Yuan Wang, Shengao Yi, Xiaojiang Li, Pengyuan Liu, Zhiwei Yang, Ronita Bardhan, Rudi Stouffs

机构 * Department of Architecture, National University of Singapore, Singapore 117566, Singapore Cambridge Centre for Advanced Research Sustainable Design Group, Department of Architecture, University of Cambridge, Cambridge, United Kingdom Department of City Regional Planning, University of Pennsylvania, Philadelphia, PA 19104, USA Urban Analytics Subject Group, Urban Studies \& Social Policy Division, University of Glasgow Laboratory for Earth Surface Processes, Ministry of Education, College of Urban Environmental Sciences, Peking University, Beijing 100871, China

AI总结 本文通过比较地表温度与通用热气候指数,揭示城市形态对人类热压力的影响,采用可解释的机器学习方法分析两者在空间分布和机制上的差异。

Comments Accepted manuscript. The final published version is available at https://doi.org/10.1016/j.scs.2026.107659

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.15727 2026-07-17 cs.CR cs.AI cs.LG cs.MA cs.SE 版本更新

AgentWorm: Self-Propagating Attacks Across LLM Agent Ecosystems

ClawWorm:针对LLM代理生态系统自主传播的攻击

Yihao Zhang, Zeming Wei, Xiaokun Luan, Chengcan Wu, Zhixin Zhang, Jiangrong Wu, Haolin Wu, Huanran Chen, Jun Sun, Meng Sun

机构 * Peking University(北京大学) Sun Yat-sen University(中山大学) Wuhan University(武汉大学) Tsinghua University(清华大学) Singapore Management University(新加坡管理学院)

AI总结 研究提出ClawWorm,首个自主传播的LLM代理框架攻击,通过单条消息实现持久化感染与多跳传播,揭示模型安全姿态差异及防御策略。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.00867 2026-07-16 cs.CV 版本更新

EFlow: Learning Evidence Flow for Long-Video Reasoning with Adaptive Reflection

EFlow: 学习证据流用于具有自适应反思的长视频推理

Wenhao Zhang, Kuanwei Lin, Xuyi Yang, Wei Gao, Ge Li

机构 * School of Electronic and Computer Engineering, Peking University(北京大学电子与计算机工程学院) The Hong Kong University of Science and Technology(香港科技大学)

AI总结 提出EFlow框架,通过分离时间定位与逻辑推理(CoT)及置信度感知的反思机制,解决长视频推理中早期语义假设导致的证据偏差问题,在五个基准上提升性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.11245 2026-07-15 cs.SE cs.AI 版本更新

An Empirical Study for Android-to-OpenHarmony GUI Test Migration

从安卓到开源鸿蒙系统的图形用户界面测试迁移实证研究

Yakun Zhang, Xinjia Chen, Yiyun Chen, Yuxia Zhang, Mingyi Zhou, Xiang Gao, Shaokun Zhang, Li Li, Yunming Ye

机构 * Harbin Institute of Technology(哈尔滨工业大学) Beijing Institute of Technology(北京理工大学) Beihang University(北航) Peking University(北京大学)

AI总结 研究从安卓到开源鸿蒙系统的图形用户界面测试迁移问题,构建数据集,选择并适配两种先进迁移方法进行评估,发现现有方法效果不佳,进而提出增强方法ITeM-HM,显著提升了测试迁移成功率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.01117 2026-07-15 cs.CV 版本更新

MoHallBench: A Benchmark for Motion Hallucination in Video Large Language Models

MoHallBench: 视频大语言模型中运动幻觉的基准测试

Sihan Chen, Jiale Li, Jianghang Lin, Mengyuan Liu

机构 * Xiamen University(厦门大学) South China University of Technology(华南理工大学) Peking University(北京大学)

AI总结 提出MoHallBench基准,系统评估视频大语言模型中的运动幻觉,涵盖共现先验、顺序推理和相似混淆三类来源,揭示动作识别与幻觉抵抗的解耦现象。

Comments 19 pages, 5 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.06486 2026-07-15 cs.AI 版本更新

JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks

JADE:面向开放式专业任务的专家基础动态评估

Lanbo Lin, Jiayao Liu, Tianyuan Yang, Li Cai, Yuanwu Xu, Lei Wei, Sicong Xie, Guannan Zhang

机构 * Alibaba International Digital Commerce Group(阿里巴巴国际数字商业集团) Zhejiang University(浙江大学) Peking University(北京大学)

AI总结 提出JADE双层评估框架,结合专家知识与动态声明级评估,解决开放式专业任务中严格性与灵活性的矛盾,在BizBench等基准上提升稳定性并揭示关键失败模式。

Comments Accepted at ICML 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2312.17670 2026-07-15 cs.CV cs.LG q-bio.QM q-bio.TO 版本更新

The TopCoW Challenge -- Topology-Aware Circle of Willis Segmentation for CT and MR Angiography

TopCoW挑战——用于CT和MR血管造影的拓扑感知Willis环分割

Kaiyuan Yang, Fabio Musio, Yihui Ma, Norman Juchler, Johannes C. Paetzold, Rami Al-Maskari, Luciano Höher, Hongwei Bran Li, Ibrahim Ethem Hamamci, Anjany Sekuboyina, Suprosanna Shit, Houjing Huang, Chinmay Prabhakar, Ezequiel de la Rosa, Bastian Wittmann, Diana Waldmannstetter, Florian Kofler, Fernando Navarro, Martin J. Menten, Ivan Ezhov, Daniel Rueckert, Iris N. Vos, Ynte M. Ruigrok, Birgitta K. Velthuis, Hugo J. Kuijf, Pengcheng Shi, Wei Liu, Ting Ma, Maximilian R. Rokuss, Yannick Kirchhoff, Fabian Isensee, Klaus Maier-Hein, Chengcheng Zhu, Huilin Zhao, Philippe Bijlenga, Julien Hämmerli, Catherine Wurster, Laura Westphal, Jeroen Bisschop, Elisa Colombo, Hakim Baazaoui, Hannah-Lea Handelsmann, Andrew Makmur, James Hallinan, Amrish Soundararajan, Benedikt Wiestler, Jan S. Kirschke, Evamaria O. Riedel, Roland Wiest, Emmanuel Montagnon, Laurent Letourneau-Guillon, Kwanseok Oh, Dahye Lee, Orhun Utku Aydin, Adam Hilbert, Jana Rieger, Dimitrios Rallios, Satoru Tanioka, Alexander Koch, Dietmar Frey, Abdul Qayyum, Moona Mazher, Steven Niederer, Nico Disch, Julius C. Holzschuh, Dominic LaBella, Francesco Galati, Daniele Falcetta, Maria A. Zuluaga, Chaolong Lin, Haoran Zhao, Zehan Zhang, Minghui Zhang, Xin You, Hanxiao Zhang, Guang-Zhong Yang, Yun Gu, Sinyoung Ra, Jongyun Hwang, Hyunjin Park, Junqiang Chen, Marek Wodzinski, Henning Müller, Nesrin Mansouri, Florent Autrusseau, Cansu Yalcin, Rachika E. Hamadache, Clara Lisazo, Joaquim Salvi, Adrià Casamitjana, Xavier Lladó, Uma Maria Lal-Trehan Estrada, Valeriia Abramova, Luca Giancardo, Arnau Oliver, Paula Casademunt, Adrian Galdran, Matteo Delucchi, Oscar Camara, Jialu Liu, Haibin Huang, Yue Cui, Zehang Lin, Yusheng Liu, Shunzhi Zhu, Tatsat R. Patel, Adnan H. Siddiqui, Vincent M. Tutino, Maysam Orouskhani, Huayu Wang, Mahmud Mossa-Basha, Yuki Sato, Sven Hirsch, Susanne Wegener, Bjoern Menze

机构 * Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland Institute of Computational Life Sciences, Zurich University of Applied Sciences (ZHAW), Waedenswil, Switzerland Department of Neuroradiology, University Hospital of Zurich, Zurich, Switzerland Department of Neurosurgery, Zhongnan Hospital of Wuhan University, Wuhan, China Department of Radiology at Weill Cornell Medicine, Cornell University, New York, USA Institute for Tissue Engineering School of Computation, Information Technology, Technical University of Munich, Germany Athinoula A. Martinos Center for Biomedical Imaging, Harvard Medical School, Boston, USA School of Medicine Health, TUM Klinikum, Technical University of Munich, Germany Munich Center for Machine Learning, Munich, Germany Department of Computing, Imperial College London, London, UK Image Sciences Institute, UMC Utrecht, Utrecht, The Netherlands Department of Neurology Neurosurgery, University Medical Center Utrecht, Utrecht, The Netherlands Department of Radiology, University Medical Center Utrecht, Utrecht, The Netherlands Electronic \& Information Engineering School, Harbin Institute of Technology (Shenzhen), China Peng Cheng Laboratory, Shenzhen, China Division of Medical Image Computing, German Cancer Research Center (DKFZ), Heidelberg, Germany Faculty of Mathematics Computer Science, Heidelberg University, Germany Helmholtz Imaging, German Cancer Research Center, Heidelberg, Germany Data Science School for Health, Karlsruhe/Heidelberg, Germany Learning Group, Department of Radiation Oncology, Heidelberg University Hospital Department of Radiology, University of Washington, Seattle, WA, USA Department of Radiology, Ren Ji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China Department of Clinical Neurosciences, Division of Neurosurgery, Geneva University Hospitals, Geneva, Switzerland Department of Neurology, University Hospital of Zurich, Zurich, Switzerland Department of Physiology, University of Toronto, Canada Department of Neurosurgery, University Hospital of Zurich, Zurich, Switzerland Department of Diagnostic Imaging, National University Hospital, Singapore University of Chicago, USA Department of Diagnostic Interventional Neuroradiology, University Hospital Berne University of Berne, Berne, Switzerland Centre de Recherche du Centre Hospitalier de l’Université de Montréal (CRCHUM), Montréal, Québec, Canada DEEPNOID Inc., Seoul, South Korea Department of Artificial Intelligence, Korea University, Seoul, South Korea Charité Lab for AI in Medicine (CLAIM), Charité Universitätsmedizin Berlin, Berlin, Germany Lung Institute, Faculty of Medicine, Imperial College London, London, UK Centre for Medical Image Computing, Department of Computer Science, University College London, London, UK Department of Radiation Oncology, Duke University Medical Center, Durham, NC, USA Institute of Medical Technology, Peking University Health Science Center, Beijing, China Hangzhou Genlight MedTech Co., Ltd., China Institute of Medical Robotics, Shanghai Jiao Tong University, Shanghai, China Department of Automation, Shanghai Jiao Tong University, Shanghai, China Department of Artificial Intelligence, Sungkyunkwan University, Seoul, South Korea Department of Electrical Computer Engineering, Sungkyunkwan University, Seoul, South Korea Shanghai MediWorks Precision Instruments Co., Ltd., China Institute of Informatics, HES-SO Valais-Wallis, Switzerland Department of Measurement Electronics, AGH University of Krakow, Poland Laboratoire de Thermique et Energie de Nantes (LTeN), Université Nantes, Polytech’Nantes, Nantes, France Research Institute of Computer Vision Center for Precision Health, McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, USA Physense, BCN-Medtech, Department of Communication Information Technologies, Universitat Pompeu Fabra, Barcelona, Spain Department of Mathematical Modeling Machine Learning, University of Zurich, Zurich, Switzerland Laboratory of Brain Atlas Brain-inspired Intelligence, Institute of Automation, Chinese Academy of Sciences, Beijing, China School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China School of Computer Information Engineering, Xiamen University of Technology, Xiamen, China Vascular Research Center, University at Buffalo, NY, USA Department of Pathology Anatomical Sciences, University at Buffalo, NY, USA Department of Neurosurgery, University at Buffalo, NY, USA LPIXEL Inc., Tokyo, Japan

AI总结 组织TopCoW基准挑战,发布含125对MRA和CTA扫描的注释数据集,参与者提交CoW分割和变体分类算法,经评估,最佳算法在多任务中表现出色,证明CoW分割算法对下游临床应用有可解释性效用。

Comments Summary paper for the TopCoW Challenge: 4 figures, 1 table, and supplementary material in appendix. Accepted for publication in NEJM AI. Datasets and best-performing algorithm Dockers are available at https://zenodo.org/records/15692630 and https://zenodo.org/records/15665435

Journal ref NEJM AI 2026;3(8)

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.06537 2026-07-14 cs.RO 版本更新

UniLM-Nav: A Unified Framework for Zero-Shot Last-Mile Navigation

UniLM-Nav:零样本最后一英里导航的统一框架

Zhuofan Zhang, Tianxu Wang, Guoxi Zhang, Yixiong Lin, Xilin Wang, Hongming Xu, Qing Li, Song-Chun Zhu, Lifeng Fan

机构 * Tsinghua University(清华大学) State Key Laboratory of General Artificial Intelligence, BIGAI(通用人工智能国家重点实验室,字节跳动公司人工智能研究院) Harbin Institute of Technology(哈尔滨工业大学) Peking University(北京大学)

AI总结 研究移动操作中最后一英里导航问题,提出UniLM-Nav统一框架,通过多模态大语言模型后端分解任务为视图选择、功能接地和姿态推理,在OVMM基准上优于现有方法,还验证了在实际机器人上的适用性。

Comments Project page: https://unilm-nav.github.io

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.29473 2026-07-14 cs.CV 版本更新

MAVIN: Multi-Shot Audio-Visual Generation with Customized Narrative Control

MAVIN:具有叙事控制的多镜头视听生成

Kaiqi Liu, Yunyao Mao, Ziqi Cai, Zheng Geng, Jing Wang, Qiulin Wang, Xintao Wang, Pengfei Wan, Kun Gai, Shuchen Weng, Boxin Shi

机构 * Peking University(北京大学) Kling Team, Kuaishou Technology(快手团队) Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所) Sun Yat-sen University(中山大学)

AI总结 提出MAVIN框架,通过边界感知注意力、ID感知传播和多智能体脚本管线,实现多镜头视听生成中的叙事控制,解决时间错位、可控性差和脚本不完整问题。

Comments Accepted to ECCV 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.02776 2026-07-14 cs.RO 版本更新

XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations

XR-1:通过学习统一的视觉-运动表示实现多功能的视觉-语言-动作模型

Shichao Fan, Kun Wu, Zhengping Che, Xinhua Wang, Di Wu, Fei Liao, Ning Liu, Yixue Zhang, Zhen Zhao, Zhiyuan Xu, Meng Li, Qingjie Liu, Shanghang Zhang, Min Wan, Jian Tang

机构 * Beijing Innovation Center of Humanoid Robotics, Beijing, China(北京人形机器人创新中心,北京,中国) School of Mechanical Engineering and Automation, Beihang University, Beijing, China(北京航空航天大学机械工程及自动化学院,北京,中国) State Key Laboratory of Virtual Reality Technology and Systems, SCSE, Beihang University, Beijing, China(虚拟现实技术与系统国家重点实验室,SCSE,北京航空航天大学,北京,中国) State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University, Beijing, China(多媒体信息处理国家重点实验室,计算机科学学院,北京大学,北京,中国)

AI总结 XR-1通过学习统一的视觉-运动表示,解决视觉-语言-动作模型在低级动作生成和跨数据源领域差距的挑战,提出三阶段训练方法并验证了其在多种机器人和任务上的优越性能。

Comments Accepted to ICML2026 as Oral

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.16112 2026-07-14 cs.CV 版本更新

AutoV: Loss-Oriented Ranking for Visual Prompt Retrieval in LVLMs

AutoV:面向视觉提示检索的损失导向排名用于大视觉-语言模型

Yuan Zhang, Chun-Kai Fan, Sicheng Yu, Junwen Pan, Tao Huang, Ming Lu, Kuan Cheng, Qi She, Shanghang Zhang

机构 * School of Computer Science, Peking University(北京大学计算机学院) ByteDance Inc.(字节跳动公司) Shanghai Jiao Tong University(上海交通大学)

AI总结 AutoV通过损失导向的提示检索提升大视觉-语言模型在图像理解等任务中的性能。

Comments Accepted by ECCV 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.04338 2026-07-14 cs.LG 版本更新

Riemannian Denoising Diffusion Probabilistic Models

黎曼去噪扩散概率模型

Zichen Liu, Wei Zhang, Christof Schütte, Tiejun Li

机构 * Center for Data Science, Peking University(数据科学中心,北京大学) Zuse Institute Berlin(柏林Zuse研究所) Institute of Mathematics, Freie Universität Berlin(柏林自由大学数学研究所) LMAM and School of Mathematical Sciences, Center for Machine Learning Research and Center for Data Science, Peking University(LMAM和数学系,机器学习研究中心和数据科学中心,北京大学)

AI总结 本文提出了一种基于投影方案的黎曼去噪扩散概率模型,适用于更一般的流形,无需复杂的几何信息,通过理论分析和实验验证展示了其在高维流形数据建模中的有效性。

Journal ref Communications in Mathematical Sciences, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.03636 2026-07-14 stat.ML cs.LG math.ST stat.AP stat.ME stat.TH 版本更新

Likelihood Matching for Diffusion Models

扩散模型的似然匹配

Lei Qian, Wu Su, Yanqi Huang, Song Xi Chen

机构 * Center for Data Science(数据科学中心) Peking University(北京大学) Guanghua School of Management(光华管理学院) Department of Statistics and Data Science(统计与数据科学系) Tsinghua University(清华大学)

AI总结 研究提出似然匹配方法训练扩散模型,通过建立目标数据分布似然与反向扩散样本路径似然的等价关系,利用拟似然近似反向转移密度,估计得分和海森矩阵函数,引入随机采样器,建立一致性并提供收敛保证,经评估验证了该方法的有效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.07372 2026-07-14 cs.CL cs.AI 版本更新

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

通过可扩展查找实现条件记忆:大语言模型稀疏性的新轴

Xin Cheng, Rui Tian, Wangding Zeng, Damai Dai, Qinyu Chen, Bingxuan Wang, Zhenda Xie, Kezhao Huang, Xingkai Yu, Chengqi Deng, Shangyan Zhou, Chenggang Zhao, Zhewen Hao, Yukun Li, Han Zhang, Zhengyan Zhang, Yixu Wei, M. Y Xu, Huishuai Zhang, Dongyan Zhao, Wenfeng Liang

机构 * Peking University(北京大学) DeepSeek-AI

AI总结 研究针对Transformer缺乏知识查找原语的问题,引入条件记忆及Engram模块,通过制定稀疏分配问题发现U形缩放定律,扩展Engram至27B参数,在多领域提升性能,揭示其优势,为下一代稀疏模型提供重要建模原语。

详情

展开后加载摘要…

URL PDF HTML 收藏
2404.03578 2026-07-14 cs.LG stat.ML 版本更新

Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithms

具有交互式数据收集的分布鲁棒强化学习:基本难度与近最优算法

Miao Lu, Han Zhong, Tong Zhang, Jose Blanchet

机构 * Department of Management Science and Engineering, Stanford University(斯坦福大学管理科学与工程系) Center for Data Science, Peking University(北京大学数据科学中心) Department of Computer Science, University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校计算机科学系)

AI总结 针对强化学习中模拟与现实的差距,提出分布鲁棒强化学习,通过交互式数据收集应对挑战。因支持转移难题,引入消失最小值假设,给出近最优算法,扩展到新公式和博弈,应用于库存控制。

详情

展开后加载摘要…

URL PDF HTML 收藏