arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

The Chinese University of Hong Kong(香港中文大学)

2026-02-04 至 2026-02-04 共收录 14
2602.03798 2026-02-04 cs.SE cs.CL cs.CV

FullStack-Agent: Enhancing Agentic Full-Stack Web Coding via Development-Oriented Testing and Repository Back-Translation

全栈代理:通过面向开发的测试和仓库反翻译增强全栈网页编码

Zimu Lu, Houxing Ren, Yunqiao Yang, Ke Wang, Zhuofan Zong, Mingjie Zhan, Hongsheng Li

机构 * Multimedia Laboratory (MMLab), The Chinese University of Hong Kong(香港中文大学多媒体实验室) Shenzhen Loop Area Institute(深圳河套学院)

AI总结 FullStack-Agent通过面向开发的测试和仓库反翻译提升全栈网页编码能力,其多代理框架和自我改进方法在多个测试用例中表现出色。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.03747 2026-02-04 cs.CV

LIVE: Long-horizon Interactive Video World Modeling

LIVE: 长时距交互视频世界建模

Junchao Huang, Ziyang Ye, Xinting Hu, Tianyu He, Guiyu Zhang, Shaoshuai Shi, Jiang Bian, Li Jiang

机构 * The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)) Shenzhen Loop Area Institute(深圳河套学院) Microsoft Research(微软研究院) The University of Hong Kong(香港大学) Voyager Research, Didi Chuxing Project(Voyager研究,滴滴出行项目)

AI总结 LIVE通过循环一致性目标限制误差累积,无需教师蒸馏,实现长时距交互视频世界建模,取得最优性能。

Comments 18 pages, 22 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.03647 2026-02-04 cs.AI cs.CL

Search-R2: Enhancing Search-Integrated Reasoning via Actor-Refiner Collaboration

Search-R2: 通过演员-细化协作提升搜索集成推理

Bowei He, Minda Hu, Zenan Xu, Hongru Wang, Licheng Zong, Yankai Chen, Chen Ma, Xue Liu, Pluto Zhou, Irwin King

机构 * The Chinese University of Hong Kong(香港中文大学) LLM Department, Tencent(腾讯语言模型部门) Mohamed bin Zayed University of Artificial Intelligence(马尔代夫人工智能大学) McGill University(麦吉尔大学) City University of Hong Kong(香港城市大学) The University of Edinburgh(爱丁堡大学)

AI总结 Search-R2通过演员-细化协作框架提升搜索集成推理,通过针对性干预和混合奖励设计,在多种QA数据集上超越强基线,实现更优的推理准确性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.03529 2026-02-04 cs.NI cs.AI cs.MM

Morphe: High-Fidelity Generative Video Streaming with Vision Foundation Model

Morphe: 基于视觉基础模型的高保真生成视频流媒体

Tianyi Gong, Zijian Cao, Zixing Zhang, Jiangkai Wu, Xinggong Zhang, Shuguang Cui, Fangxin Wang

机构 * School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen(科学与工程学院,香港中文大学(深圳)) Shenzhen Future Network of Intelligence Institute(深圳未来网络智能研究院) Wangxuan Institute of Computer Technology, Peking University(王轩计算机技术研究所,北京大学)

AI总结 Morphe基于视觉基础模型,通过联合训练和智能分组丢弃技术,实现高保真度、低带宽的实时视频流媒体传输。

Comments Accepted by NSDI 2026 Fall

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.03320 2026-02-04 cs.CV cs.AI

MedSAM-Agent: Empowering Interactive Medical Image Segmentation with Multi-turn Agentic Reinforcement Learning

MedSAM-Agent: 通过多轮代理强化学习赋能交互式医学图像分割

Shengyuan Liu, Liuxin Bao, Qi Yang, Wanting Geng, Boyun Zheng, Chenxin Li, Wenting Chen, Houwen Peng, Yixuan Yuan

机构 * Chinese University of Hong Kong, Hong Kong SAR, China(香港中文大学) Hunyuan Group, Tencent(腾讯洪音集团) Institute of Automation, the Chinese Academy of Sciences, Beijing, China(中国科学院自动化研究所) Dalian University of Technology, Dalian, China(大连理工大学) Stanford University, Stanford, USA(斯坦福大学)

AI总结 MedSAM-Agent通过多轮代理强化学习提升医学图像分割的交互效率与准确性

Comments 23 Pages, 4 Figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.03268 2026-02-04 cs.LG cs.AI cs.MM

Unveiling Covert Toxicity in Multimodal Data via Toxicity Association Graphs: A Graph-Based Metric and Interpretable Detection Framework

通过毒性关联图揭示多模态数据中的隐性毒性:一种基于图的度量方法和可解释检测框架

Guanzong Wu, Zihao Zhu, Siwei Lyu, Baoyuan Wu

机构 * The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)) State University of New York at Buffalo(纽约州立大学布法罗分校)

AI总结 本文提出基于图的毒性关联图方法,通过多模态毒性隐性度度量标准实现对隐性毒性的可解释检测,提升多模态毒性检测的透明度和准确性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.02927 2026-02-04 stat.ML cs.LG

Training-Free Self-Correction for Multimodal Masked Diffusion Models

无需训练的多模态掩码扩散模型自校正

Yidong Ouyang, Panwen Hu, Zhengyan Wan, Zhe Wang, Liyan Xie, Dmitriy Bespalov, Ying Nian Wu, Guang Cheng, Hongyuan Zha, Qiang Sun

机构 * University of California, Los Angeles(加州大学洛杉矶分校) Mohamed bin Zayed University of Artificial Intelligence(莫莫德·本·扎耶德人工智能大学) East China Normal University(华东师范大学) University of Virginia(弗吉尼亚大学) University of Minnesota(明尼苏达大学) Drexel university(德雷塞尔大学) The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)) University of Toronto(多伦多大学)

AI总结 本文提出无需训练的多模态掩码扩散模型自校正方法,通过减少采样步骤提升生成质量,适用于文本到图像和多模态理解任务。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.02567 2026-02-04 cs.LG cs.AI eess.IV

IceBench-S2S: A Benchmark of Deep Learning for Challenging Subseasonal-to-Seasonal Daily Arctic Sea Ice Forecasting in Deep Latent Space

IceBench-S2S:一种深度学习用于挑战性亚季节至季节每日北极海冰预测的基准

Jingyi Xu, Shengnan Wang, Weidong Yang, Siwei Tu, Lei Bai, Ben Fei

机构 * Fudan University(复旦大学) Shanghai AI Laboratory(上海人工智能实验室) Chinese University of Hong Kong(香港中文大学)

AI总结 IceBench-S2S通过深度学习框架提升北极海冰预测的季节尺度能力,为极地环境监测提供统一的训练和评估流程。

Comments 9 pages, 6 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.02550 2026-02-04 cs.LG cs.AI

HyPAC: Cost-Efficient LLMs-Human Hybrid Annotation with PAC Error Guarantees

HyPAC: 低成本的LLM-人类混合标注与PAC误差保证

Hao Zeng, Huipeng Huang, Xinhao Qu, Jianguo Huang, Bingyi Jing, Hongxin Wei

机构 * Department of Statistics and Data Science, Southern University of Science and Technology, China(统计与数据科学系,南方科技大学) Department of Statistics, University of California at Riverside, USA(统计系,加州大学河滨分校) College of Computing and Data Science, Nanyang Technological University, Singapore(计算与数据科学学院,南洋理工大学) School of Artificial Intelligence, The Chinese University of Hong Kong, Shenzhen, China(人工智能学院,香港中文大学(深圳))

AI总结 HyPAC通过自适应路由和PAC误差保证,实现低成本高效标注,降低78.51%成本并严格控制误差。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.01467 2026-02-04 quant-ph cs.IT cs.LG math.IT

Quantum Information Ordering and Differential Privacy

量子信息排序与差分隐私

Naqueeb Ahmad Warsi, Ayanava Dasgupta, Masahito Hayashi

机构 * Indian Statistical Institute(印度统计研究所) School of Data Science, The Chinese University of Hong Kong, Shenzhen(香港中文大学深圳校区数据科学学院) International Quantum Academy(国际量子学院) Graduate School of Mathematics, Nagoya University(名古屋大学数学研究生院)

AI总结 本文提出通过量子态信息性顺序定义量子差分隐私,研究隐私化假设检验和参数估计的限制,并建立量子通道的收缩界。

Comments 36 pages, 2 figures; Significant revision: This manuscript has been restructured to focus exclusively on Quantum Information Ordering and Privacy definitions. The results regarding Stability, which appeared in earlier versions of this preprint, have been moved to a separate companion paper: arXiv:2602.01177

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.00060 2026-02-04 cs.RO

Correspondence-Free, Function-Based Sim-to-Real Learning for Deformable Surface Control

无需对应的功能基仿真到现实学习用于变形表面控制

Yingjun Tian, Guoxin Fang, Renbo Su, Aoran Lyu, Neelotpal Dutta, Weiming Wang, Simeon Gill, Andrew Weightman, Charlie C. L. Wang

机构 * Department of Mechanical and Aerospace Engineering, The University of Manchester, United Kingdom(机械与航空航天工程系,曼彻斯特大学,英国) Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong(机械与自动化工程系,香港中文大学(深圳),香港) Department of Materials, The University of Manchester, United Kingdom(材料系,曼彻斯特大学,英国)

AI总结 本文提出一种无需对应的功能基仿真到现实学习方法,用于控制可变形自由形式表面,通过神经网络参数化变形函数空间和置信度图,实现无需对应关系的仿真到现实转移。

Comments arXiv admin note: text overlap with arXiv:2405.08935

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.23729 2026-02-04 cs.CV

Proteus-ID: ID-Consistent and Motion-Coherent Video Customization

Proteus-ID:身份一致且运动协调的视频定制

Guiyu Zhang, Chen Shi, Zijian Jiang, Xunzhi Xiang, Jingjing Qian, Shaoshuai Shi, Li Jiang

机构 * The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)) Nanjing University(南京大学) Voyager Research, Didi Chuxing(Voyager Research与滴滴出行)

AI总结 Proteus-ID通过多模态身份融合、时间感知身份注入和自适应运动学习,实现了身份一致且运动协调的视频定制,提升了视频生成的真实性和流畅度。

Comments SIGGRAPH Asia 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.17873 2026-02-04 cs.CV cs.AI

SurgVidLM: Towards Multi-grained Surgical Video Understanding with Large Language Model

SurgVidLM:迈向多粒度外科视频理解的大型语言模型

Guankun Wang, Junyi Wang, Wenjin Mo, Long Bai, Kun Yuan, Ming Hu, Jinlin Wu, Junjun He, Yiming Huang, Nicolas Padoy, Zhen Lei, Hongbin Liu, Nassir Navab, Hongliang Ren

机构 * The Chinese University of Hong Kong(香港中文大学) Sun Yat-sen University(中山大学) University of Strasbourg(斯特拉斯堡大学) Technical University of Munich(慕尼黑技术大学) Monash University(墨尔本大学) Centre for Artificial Intelligence and Robotics, HKISI-CAS(人工智能与机器人中心,HKISI-CAS) Shanghai AI Laboratory(上海人工智能实验室)

AI总结 SurgVidLM通过多粒度分析提升外科视频理解能力,结合全局与局部机制实现更精确的手术流程解析。

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.00592 2026-02-04 cs.LG cs.AI cs.ET

Hyper-Compression: Model Compression via Hyperfunction

超压缩:通过超函数进行模型压缩

Fenglei Fan, Juntong Fan, Dayang Wang, Jingbo Zhang, Zelin Dong, Shijun Zhang, Ge Wang, Tieyong Zeng

机构 * Department of Data Science, City University of Hong Kong, China SAR(数据科学系,香港城市大学,中国香港特别行政区) Department of Applied Mathematics, The Hong Kong Polytechnic University, China SAR(应用数学系,香港理工大学,中国香港特别行政区) Department of Biomedical Engineering, Rensselaer Polytechnic Institute, NY, US(生物医学工程系,罗切斯特理工学院,纽约,美国) Department of Mathematics, The Chinese University of Hong Kong, China SAR(数学系,香港中文大学,中国香港特别行政区)

AI总结 通过超函数实现模型压缩,无需重新训练,具有高压缩比和低推理时间的特性。

详情

展开后加载摘要…

URL PDF HTML 收藏