Sizhe Chen (陈思哲)

Hi! I am a CS Ph.D. candidate at UC Berkeley in Berkeley AI Research (BAIR), working with Prof. David Wagner. I study real-world AI security problems with general and principled solutions. I led the development of StruQ, SecAlign, Meta-SecAlign, and SecOPD, the most robust open LLMs against adaptive prompt injections, the top-1 threat to AI agents. My research has been supported by selective industry fundings, including NVIDIA Graduate Fellowship, Meta FAIR and Google DeepMind through BAIR Commons. I obtained an M.Eng. and a B.Eng. from Shanghai Jiao Tong University.

Securing LLMs Against Prompt Injection for Agentic Applications