Our paper ''LAMDA: Leveraging Large Language Models for Better API Misuse Detection via Multi-Agent Collaboration'' has been accepted by TOSEM.
Peking University
Code Security Auditing Expert
I am a third-year Ph.D. student at Peking University, advised by Prof. Dan Hao and Prof. Zeyu Sun. Before joining Peking University, I received my Master's degree in Electronic Information (Software Engineering) from the School of Computer Science, Wuhan University, under the supervision of Prof. Jin Liu.
My research focuses on training LLM for code security auditing. I work on building security-aware LLMs that can understand program semantics, reason about vulnerable behaviors, and support automated vulnerability discovery and repair. My recent work has been to train LLMs as practical auditing agents for complex software systems.
Our paper ''LAMDA: Leveraging Large Language Models for Better API Misuse Detection via Multi-Agent Collaboration'' has been accepted by TOSEM.
Our paper ''CausalRepair: Bridging the Causality Gap in Large Language Model-based Automated Program Repair via Dual-Slicing'' has been accepted by ISSTA 2026.
Our paper ''Search-Induced Issues in Web-Augmented LLM Code Generation: Detecting and Repairing Error-Inducing Pages'' has been accepted by ISSTA 2026.
Our paper ''Decomposing Ponzi Schemes: Multi-aspect Lifecycle Analysis for Detecting Fraudulent Smart Contracts'' has been accepted by TOSEM.
Our paper ''Decoupling Code Classification for Semantic Transfer: Learning from LLM-Generated Explanations'' has been accepted by TOSEM.
Open to OpportunitiesI am currently on the academic/industrial job market for research positions. Please feel free to reach out for potential collaborations!