Jiawei Liu 刘嘉威

Jiawei Liu 刘嘉威

Associate Professor
Department of Automation
University of Science and Technology of China

About Me

Hi! I’m Jiawei Liu (刘嘉威), an Associate Professor in the Department of Automation at the University of Science and Technology of China (中国科学技术大学). I work with Prof. Zheng-Jun Zha at the Lab for Multimodal Embodied Intelligence (多模态具身智能实验室). My research focuses on human-centric visual understanding, fake media content analysis, multimodal large language models, and embodied AI. I have published more than 60 papers in leading AI journals and conferences.

I received my B.S. degree from Hefei University of Technology (HFUT) in 2013 and my Ph.D. degree from the University of Science and Technology of China (USTC) in 2019. During my Ph.D., I was a research intern at Microsoft Research Asia (MSRA) under the supervision of Prof. Tao Mei from October 2016 to January 2017. From September 2025 to August 2026, I was a visiting scholar at the National University of Singapore (NUS) , under the supervision of Prof. Shuicheng Yan.

I welcome applications from motivated graduate students interested in computer vision, multimedia analysis, and embodied AI. Students with backgrounds in mathematics, machine learning, or programming are encouraged to email me a brief introduction and CV. I also welcome opportunities for academic collaboration. Please contact me at jwliu6@ustc.edu.cn.

🌟 欢迎热爱科研、愿意主动投入的本科生加入课题组，参与科研实习、大创或暑研。研究方向包括以人为中心的视觉理解、多模态大模型和具身智能。数学、物理、计算机等相关专业的同学均可通过邮件联系，并附上个人简历；有意继续深造的同学也欢迎交流。

News

  • [2026.09]:  🎉🎉 One paper on tool-augmented scientific reasoning is accepted by NeurIPS 2026.
  • [2026.09]:  🎉🎉 One paper on open-vocabulary HOI detection is accepted by NeurIPS 2026.
  • [2026.09]:  🎉🎉 One paper on point cloud segmentation is accepted by NeurIPS 2026.
  • [2026.09]:  🎉🎉 One paper on thinking with images is accepted by NeurIPS 2026.
  • [2026.09]:  🎉🎉 One paper on image forgery detection and localization is accepted by IEEE TIP 2026.
  • [2026.08]:  🎉🎉 One paper on open-ended agent RL is accepted by EMNLP Findings 2026.
  • [2026.07]:  🎉🎉 One paper on wu dialect vowel recognition is accepted by CICAI 2026.
  • [2026.07]:  🎉🎉 One paper on stealthy backdoor is accepted by IEEE TPAMI 2026.
  • [2026.06]:  🎉🎉 One paper on agentic tool use is accepted by ECCV 2026.
  • [2026.06]:  🎉🎉 One paper on visible-infrared person identification is accepted by IEEE TIFS 2026.
  • [2026.06]:  🎓🎓 Congratulations to Jingyi Xie (谢璟毅) on earning his Ph.D. degree!
  • [2026.06]:  🎓🎓 Congratulations to Changrong Hu (胡长荣) on earning her Master’s degree!
  • [2026.05]:  🎉🎉 One paper on open-ended agents is accepted by ICML 2026.
  • [2026.04]:  🎉🎉 One paper on automatic rigging is accepted by ICIC 2026. Oral
  • [2026.03]:  🏆🏆 Won the champion of the 2nd multimodal deception detection competition @CVPR 2026.
  • [2026.03]:  🏛️🏛️ Promoted to Associate Professor (tenured) at USTC.
  • [2026.02]:  🎉🎉 One paper on open-vocabulary HOI detection is accepted by CVPR 2026. Highlight
  • [2026.01]:  🎉🎉 One paper on active prompt learning is accepted by ICLR 2026.
  • [2026.01]:  🎉🎉 One paper on multimodal fact-checking is accepted by WWW 2026. Oral
  • [2025.12]:  🎉🎉 One paper on class incremental learning is accepted by CVM 2026.
  • [2025.11]:  🎉🎉 One paper on zero-shot HOI detection is accepted by IJCV 2025.
  • [2025.11]:  🎉🎉 One paper on occluded person re-identification is accepted by JICN 2025.
  • [2025.10]:  🎉🎉 One paper on active prompt learning is accepted by IJCV 2025.
  • [2025.09]:  🎉🎉 One paper on explainable video misinformation detection is accepted by NeurIPS 2025.
  • [2025.09]:  🎉🎉 One paper on multimodal knowledge editing is accepted by NeurIPS 2025.
  • [2025.09]:  🎉🎉 One paper on fake news detection is accepted by IEEE TKDE 2025.
  • [2025.09]:  ✈️✈️ Started as a visiting scholar at the National University of Singapore.
  • [2025.07]:  🎉🎉 One paper on person re-identification is accepted by ACM MM 2025.
  • [2025.06]:  🎓🎓 Congratulations to Wei Wu (吴蔚) on earning her Ph.D. degree!
  • [2025.06]:  🎓🎓 Congratulations to Hao Tong (童浩) on earning his Master’s degree!
  • [2025.04]:  🎉🎉 One paper on image forgery detection and localization is accepted by IJCAI 2025.
  • [2025.02]:  🎉🎉 One paper on test-time adaptation is accepted by CVPR 2025.
  • [2025.02]:  🎉🎉 One paper on comprehensive image captioning is accepted by CVPR 2025.
  • [2025.01]:  🎉🎉 One paper on person re-identification is accepted by IEEE TIFS 2025.
  • [2024.12]:  🎓🎓 Congratulations to Bingyu Hu (胡冰玉) on earning her Ph.D. degree!
  • [2024.12]:  🎉🎉 One paper on person re-identification is accepted by CVM 2025.
  • [2024.12]:  🎉🎉 One paper on human-object interaction is accepted by AAAI 2025.
  • [2024.10]:  🎉🎉 One paper on person re-identification is accepted by ACM TOMM 2024.
  • [2024.08]:  🎉🎉 One paper on image forgery localization is accepted by ECCV 2024.
  • [2024.06]:  🎓🎓 Congratulations to Wenjun Gan (甘文君) on earning her Master’s degree!
  • [2024.04]:  🎉🎉 One paper on person re-identification is accepted by ICME 2024.

Publications

Human-Centric Visual Understanding

Fake Media Content Analysis

Multimodal Large Language Models

Patents

  • 2026.04 一种基于强化学习框架的开放词汇人物交互检测方法查正军; 刘嘉威; 徐永超; 王浚沣发明专利专利号：CN202610429538.7
  • 2026.03 基于检索与推理增强大语言模型的多模态事实核查方法查正军; 刘嘉威; 展宏民; 张凡瑞发明专利专利号：CN202610248544.2
  • 2025.11 基于冲突感知检索和多级协同引导的多模态知识编辑方法查正军; 刘嘉威; 张强; 张凡瑞发明专利专利号：CN202511709848.6
  • 2025.10 基于双重不确定性引导特征对齐的文本到图像行人检索方法查正军; 刘嘉威; 郑雨霏发明专利专利号：CN202511451428.2
  • 2025.03 一种基于Mamba架构的人物交互检测方法查正军; 刘嘉威; 徐永超发明专利专利号：CN202510261091.2
  • 2024.10 一种基于实体增强和立场检查的多模态事实核查方法查正军; 刘嘉威; 张凡瑞; 张强发明专利专利号：CN202410403673.5
  • 2024.10 基于特征多样化和域自适应特征选择的鲁棒行人重识别方法刘嘉威; 赵国治; 吴勇; 贺强; 胡冰玉发明专利专利号：CN202410070798.0
  • 2024.10 一种基于跨模态语义对齐的文本行人重识别方法刘嘉威; 赵国治; 吴勇; 贺强; 甘文君发明专利专利号：CN202410072896.8
  • 2024.10 一种基于噪声辅助提示学习的图像伪造检测定位方法查正军; 傅雪阳; 李东; 朱佳莹; 刘嘉威发明专利专利号：CN202411253779.8
  • 2023.10 基于边缘感知的区域消息传递控制的图像伪造定位方法查正军; 傅雪阳; 李东; 朱佳莹; 刘嘉威发明专利专利号：CN202310346607.4
  • 2023.11 一种基于多层次语义增强的多模态假新闻检测方法查正军; 刘嘉威; 张强; 张凡瑞发明专利专利号：CN202311298800.1
  • 2022.10 一种基于特征对齐的退化环境下的行人重识别方法查正军; 刘嘉威; 王堃宇发明专利专利号：CN202210792619.5
  • 2022.10 基于时序补偿引导的强化学习图像-视频行人重识别方法查正军; 刘嘉威; 吴蔚发明专利专利号：CN202210362412.4
  • 2022.10 基于高斯过程的去偏批量归一化的鲁棒行人重识别方法查正军; 刘嘉威; 黄志鹏发明专利专利号：CN202210193826.9
  • 2022.10 基于模态自适应混合和不变性卷积分解的行人重识别方法查正军; 刘嘉威; 黄志鹏发明专利专利号：CN202210155715.9
  • 2020.10 检索目标的方法和装置刘武; 刘嘉威; 梅涛; 郑可成发明专利专利号：CN202010215923.4
  • 2019.10 行人再识别数据生成方法查正军; 刘嘉威发明专利专利号：CN201910466234.8

Honors & Awards

  • 2026.06 CVPR2026 Workshop: MultiModal Deception Detection Challenge (1st Place Award)
  • 2025.10 First-Class Award of Anhui Provincial Natural Science Award (3/3)
  • 2025.10 PRCV2025 Workshop: Omni Multi-modal Person Re-identification Challenge (Excellence Award)
  • 2024.07 ACM MM2024 Workshop: Micro-Action Analysis Grand Challenge (3rd Place Award)
  • 2023.03 AAAI2023 Workshop: Defactify 2: Multimodal Fake News Detection (4th Place Award)
  • 2022.01 Mozi Distinguished Youth Special Allowance of USTC
  • 2020.01 Special Research Assistantship of CAS
  • 2019.06 CAS Presidential Scholarship
  • 2019.06 Outstanding Graduates Scholarship, USTC
  • 2018.09 Individual Scholarship, USTC
  • 2016.09 Individual Scholarship, USTC

Experience

Professional Experience

  1. 2026.05 – Present
    University of Science and Technology of China Associate Professor Department of Automation Collaborating with Prof. Zheng-Jun Zha
  2. 2025.09 – 2026.08
    National University of Singapore Visiting Scholar School of Computing · Department of Computer Science Host: Prof. Shuicheng Yan
  3. 2021.12 – 2026.04
    University of Science and Technology of China Associate Research Fellow Department of Automation Collaborating with Prof. Zheng-Jun Zha
  4. 2019.10 – 2021.11
    University of Science and Technology of China Postdoctoral Research Fellow Department of Automation Collaborating with Prof. Zheng-Jun Zha and Prof. Yongdong Zhang
  5. 2019.08 – 2019.10
    JD AI Research Institute Researcher Working with Prof. Tao Mei and Prof. Wu Liu
  6. 2016.10 – 2017.01
    Microsoft Research Asia Research Intern Supervisor: Prof. Tao Mei

Education

  1. 2014.09 – 2019.06
    University of Science and Technology of China Ph.D. Control Science and Engineering · Department of Automation Advisor: Prof. Zheng-Jun Zha
  2. 2009.08 – 2013.07
    Hefei University of Technology B.S. Automation · Department of Automation Advisor: Prof. Chenghui Zhu

Funding

  • 2025.01 – 2028.12 National Natural Science Foundation of China, General ProgramPI基金委面上基金项目, 主持
  • 2025.09 – 2026.08 CCF-1688 Yuanbao Cooperation FundPICCF-1688源宝合作基金项目，主持
  • 2025.01 – 2026.12 USTC Research Funds of the Double First-Class InitiativePI校青年创新重点基金项目, 主持
  • 2022.01 – 2024.12 National Natural Science Foundation of China, Youth ProgramPI基金委青年基金项目, 主持
  • 2023.06 – 2024.12 China Merchants Bank-USTC Laboratory Collaborative ProjectPI招商银行-中国科大实验室合作项目, 主持
  • 2021.01 – 2022.12 USTC Research Funds of the Double First-Class InitiativePI校青年创新基金项目, 主持
  • 2021.01 – 2022.12 China Postdoctoral Science FoundationPI中国博士后面上基金项目, 主持
  • 2024.07 – 2029.06 Chinese Academy of Sciences Stable Support Basic Research Youth Team ProgramSubproject PI中国科学院稳定支持基础研究领域青年团队计划, 课题主持
  • 2024.04 – 2030.12 Strategic Priority Project of the Chinese Academy of Sciences on Securing the High Ground in Science and TechnologyKey Participant中国科学院抢占科技制高点专项课题，参与
  • 2025.01 – 2026.12 USTC Research Funds of the Double First-Class InitiativeKey Participant校融合专项基金项目, 参与
  • 2020.11 – 2023.10 National Key R&D Program of ChinaKey Participant科技创新2030-“新一代人工智能”重大项目, 参与
  • 2020.01 – 2023.12 Joint Funds of the National Natural Science Foundation of ChinaKey Participant基金委联合重点基金项目, 参与
  • 2017.01 – 2021.12 National Key R&D Program of ChinaKey Participant国家重点研发计划课题, 参与

Teaching

Graduate Courses

Computer Vision (计算机视觉) 60 teaching hours · Fall 2022–2026

Undergraduate Courses

Python and Deep Learning Fundamentals (Python与深度学习基础) 40 teaching hours · Spring 2025

Academic Service

Expert Service

  • Expert in the National Graduate Education Evaluation and Monitoring Expert Database
  • Expert in the National Undergraduate Thesis (Design) Random Inspection and Evaluation Expert Database
  • Member of CSIG Technical Committee on Multimedia
  • Member of CSIG Technical Committee on Digital Media Forensics and Security
  • Member of CCF Technical Committee on Multimedia Technology

Young Editorial Board Member

  • Journal of Intelligent Computing and Networking

Journal Reviewer

  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • International Journal of Computer Vision
  • IEEE Transactions on Multimedia
  • IEEE Transactions on Cybernetics
  • IEEE Transactions on Image Processing
  • IEEE Transactions on Emerging Topics in Computational Intelligence
  • IEEE Transactions on Neural Networks and Learning Systems
  • IEEE Transactions on Circuits and Systems for Video Technology
  • IEEE Transactions on Dependable and Secure Computing
  • IEEE Transactions on Artificial Intelligence
  • IEEE Transactions on Industrial Informatics
  • IEEE Transactions on Computational Social Systems
  • IEEE Transactions on Automation Science and Engineering
  • IEEE Transactions on Consumer Electronics
  • ACM Transactions on Data Science
  • ACM Transactions on Multimedia Computing Communications and Applications
  • Neurocomputing
  • Neural Networks
  • Image and Vision Computing
  • Computer Vision and Image Understanding
  • IEEE Internet of Things Journal
  • IEEE Signal Processing Letters
  • Pattern Recognition Letters
  • IEEE Sensors Letters
  • Journal of Computer Science and Technology

Conference Area Chair or Senior Program Committee

  • AAAI Conference on Artificial Intelligence
  • Chinese Control Conference

Conference Reviewer

  • NeurIPS, CVPR, ICCV, ECCV, ICML, ICLR, ACM MM, IJCAI, AAAI, ICME, PRCV, ACCV