Greetings and Welcome to My Homepage

Biography

I am currently a Research Fellow at Nanyang Technological University, working with Prof. Dacheng Tao and previously with Prof. Tianwei Zhang. Prior to this, I was a tenure-track Research Professor at Zhejiang University, where I received the Outstanding Junior Faculty Award (Qizhen Outstanding Young Scholar). I received my Ph.D. degree with honors in Computer Science and Technology from Tsinghua University, advised by Prof. Yong Jiang and Prof. Shu-Tao Xia, and my B.S. degree with honors in Applied Mathematics from Ningbo University (Yangming Innovation Class), where I was advised by Prof. Lifeng Xi. During my Ph.D., I also collaborated closely with Dr. Zhifeng Li at Tencent and Prof. Bo Li, then at UIUC.

My research centers on Responsible AI and Technical AI Governance, especially AI Risk Management and AI Copyright Protection. My long-term goal is to advance the responsible and sustainable development of AI models and systems, ultimately benefiting both society and humanity. Building on my work on specialized models, my current research increasingly focuses on Trustworthy Generative and Autonomous AI, including LLMs, diffusion models and AI agents, with embodied AI as an emerging frontier. Alongside this work, I am exploring AI-driven social simulation and the sociotechnical dimensions of AI trustworthiness, including how users perceive, interpret and respond to the security and authenticity signals produced by technical systems.

I seek simple yet effective methods grounded in deep insights and theoretical support, spanning proactive risk sensing, full-stack risk mitigation and verifiable ownership attribution. I also place strong emphasis on translating research into real-world impact, through the industrial deployment of auditing and defense tools, contributions to AI standards, and engagement with AI governance.

Call for Papers: We welcome submissions to our IEEE TPAMI Special Issue on Trustworthy AI in Pattern Recognition and Computer Vision: Attacks, Robustness, Safety, and Beyond. Submissions are now open and are handled on a rolling basis. Original research articles, extended conference papers, and survey articles within the scope of the special issue are welcome. The submission deadline is 1 March 2027. [View CFP]

Selected Research

AI has advanced rapidly in recent years, evolving from specialized models designed for specific tasks to generative models capable of performing a wide range of tasks, and increasingly toward digital agents that autonomously complete tasks and embodied agents that interact and collaborate in the physical world. A major driver of this progress has been scaling: larger models trained on larger datasets with more compute. However, capability does not imply safety, and the rush to scale has sharply increased the risk of infringement. These challenges lie at the intersection of cybersecurity and AI and cannot be solved by scaling alone. Addressing them is essential to enabling trustworthy deployment and sustaining responsible innovation.

My research advances Responsible AI Development through two interconnected pillars: AI Risk Management and AI Copyright Protection. Together, these directions connect the understanding of emerging threats with practical safeguards and reliable evidence of ownership and reuse. Building on these foundations, my long-term vision is to develop trustworthy autonomous AI systems that are secure and IP-protective by design.

• AI Risk Management
  • Risk Sensing: [S&P’26, NDSS’26 (a), NDSS’26 (b), USENIX Security’26, TDSC’26, PR’26, S&P’25, USENIX Security’25, NeurIPS’25, ICLR’25, TIFS’25, TDSC’25, CVPR’24, TIFS’24 (a), ICCV’23, TIFS’24 (b), ICLR’23, AAAI’23, PR’23, ICLR’22, ICASSP’21, ICCV’21, ICLR’21, ECCV’20 (Oral)]
  • Risk Mitigation: [USENIX Security’26, TIFS’26, NeurIPS’25, ICLR’25 (a), ICLR’25 (b), ICML’25, ACL’25 (a), ACL’25 (b), ICML’24 (a), ICML’24 (b), CVPR’25, ICLR’24 (a) (Spotlight), ICLR’24 (b), IJCAI’24, IJCV’24, NeurIPS’23, ICLR’23, ICLR’22, PR’22]
• AI Copyright Protection
  • Model Copyright: [WWW’26 (Oral), ICML’26, IJCV’26, TIFS’26, NDSS’25, CVPR’25, ICLR’25, TPAMI’25, ECCV’24, ICCV’23, AAAI’22]
  • Data Copyright: [S&P’26, CCS’26, NeurIPS’26, WWW’26, TPAMI’26 (a), TPAMI’26 (b), TIFS’26, USENIX Security’25, S&P’25 (a), S&P’25 (b), TPAMI’25, TIFS’25, NeurIPS’24, TIFS’24, TIFS’23 (TOP-25 Papers), NeurIPS’23, PR’22, NeurIPS’22 (Oral)]

News

Useful Resources

AI Security Leaderboard for Frontier Models' Safeguards, as well as its Technical Report

Stakeholder-Centric Prompt Injection Benchmark for Real-world Web Agents, as well as its Toolbox

Perspective: Rethinking Data Protection in the (Generative) Artificial Intelligence Era

Survey and Benchmark for Fingerprinting-based LLM's Copyright Auditing, as well as its Toolbox

Survey and Benchmark for for Dataset Auditing, as well as its Toolbox

Survey, Toolbox, and GitHub Resources Repo about Backdoor Learning.

A Brief Bio: Dr. Yiming Li is currently a Research Fellow at Nanyang Technological University. He was previously a tenure-track Research Professor at Zhejiang University, where he received the Outstanding Junior Faculty Award, and he obtained his Ph.D. (with honors) in Computer Science and Technology from Tsinghua University. His research centers on Responsible AI and Technical AI Governance, with an emphasis on AI security and safety, AI copyright protection, and the human-centered dimensions of AI trustworthiness. His work has appeared in leading venues including IEEE S&P, USENIX Security, CCS, NDSS, ICML, NeurIPS, ICLR, CVPR, IEEE TPAMI, IJCV, IEEE TIFS, and IEEE TDSC, and has been covered by MIT Technology Review, IEEE Spectrum, and WIRED. He co-drafted a mandatory national standard on AI safety and copyright, and his AI security algorithms have been deployed in the production systems of leading technology companies. His methods for AI agent security and video generation security won first place in competitions held at NeurIPS and ECCV. He serves as Guest Editor for IEEE TPAMI, Consulting Area Editor for IEEE TIFS, Associate Editor for Pattern Recognition, Area Chair for CVPR, ICLR, ICML, NeurIPS, and KDD, and Program Committee Member for IEEE S&P, CCS, and USENIX Security. His honors include the Best Paper Award at PAKDD, the Rising Star Award at WAIC, the KAUST Rising Stars in AI, the Elsevier–Stanford World’s Top 2% Scientists list (2024/2025/2026), and the Forbes 30 Under 30 Asia List.