I am a second-year master’s student with the State Key Laboratory of Blockchain and Data Security, Zhejiang University. I am fortunate to be advised by Prof. Zhan Qin. I also work closely with Dr. Hongwei Yao at City University of Hong Kong and Dr. Yiming Li at Nanyang Technological University. I received my B.Eng. degree in Information Security from Zhejiang University in 2025.
My research interests primarily focus on Trustworthy AI. My previous work mainly addresses AI copyright protection, especially model ownership verification via fingerprinting. Currently, I am more focused on LLM agent security, particularly the attack and defense of indirect prompt injection and automated red-teaming.
Outside research, I write things down as I learn: course notes live on my notes site and Yuque (both in Chinese), and everything else goes here.
Publications
-
AttriGuard: Defeating Indirect Prompt Injection in LLM Agents via Causal Attribution of Tool Invocations
-
FIT-Print: Towards False-claim-resistant Model Ownership Verification via Targeted Fingerprint
Awards
All awards →-
Outstanding Graduate Student, Five-Good Graduate Student, and Outstanding Graduate Student Leader
2025–26
-
Outstanding Graduate and Outstanding Undergraduate Thesis, Zhejiang University
2025
-
National Scholarship for Undergraduates
2023–24
-
First-Class Scholarship, Zhejiang University
2023–24
-
Outstanding Student, Zhejiang University
2023–24
-
Higgs Extracurricular Innovation Scholarship Second Prize
2023–24
Writing
No posts yet.