About
I recently completed my MSc at the University of Edinburgh with Distinction and am a Research Intern at TSG Lab.
Prior to this, I obtained a BSc degree in Information Management and Information System from Hefei University of Technology.
My current research focuses on AI safety, with a particular interest in agent safety and interpretability.
Selected Publications
View All →SentinelNet: Safeguarding Multi-Agent Collaboration Through Credit-Based Dynamic Threat Detection
Yang Feng, Xudong Pan
Proceedings of the ACM Web Conference 2026
News
Completed my MSc in Operational Research with Data Science at the University of Edinburgh with Distinction.
Our paper "PEAR: Permutation-Equivariant Adaptive Routing Multi-Agent Debate" was accepted as a poster at EMNLP 2026 Main Conference.
Starting as a Research Intern at TSG Lab, working on AI interpretability.
Our paper "SentinelNet: Safeguarding Multi-Agent Collaboration Through Credit-Based Dynamic Threat Detection" was accepted as a poster at The Web Conference (WWW) 2026.
Starting my Master's at University of Edinburgh.
Starting as a Research Assistant at the International Research Institute of Global Cyberspace Governance, Fudan University, working on LLM-based agent vulnerabilities and defense mechanisms.