Portrait of Sizhe Chen at a concert

Sizhe Chen (陈思哲)

AI Security Researcher

Scholar GitHub LinkedIn X CV Xiaohongshu

Hi! I am a CS Ph.D. candidate at UC Berkeley AI Research (BAIR) with David Wagner (also closely with Chuan Guo and Sewon Min). I study real-world AI security problems with general and principled solutions. I led research towards the most secure open LLMs against prompt injection, the top-1 threat to AI agents. Featured work includes StruQ, (Meta-)SecAlign, SecOPD, and SecPO (check out our latest Qwen3.6-27B-Instruct-SecPO that is robust to the current strongest adaptive attacks). 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

Feel free to email me for a talk (abstract and slides) or lecture (slides).

UC Berkeley Security Seminar2027
Stanford Security Lunch2026
Princeton PLI Seminar2026
UC Berkeley Deep Learning Guest Lecture2026
ByteDance Agent Security Talk2026
NVIDIA Agent & Security Talk2026
UIUC Security Seminar2026
IBM Research Lunch Talk2026
Meta FAIR Security and Privacy Group Seminar2026
UC Berkeley NLP Group Seminar2026
UC Berkeley Computer Security Guest Lecture2026
Shanghai AI Lab Xinghe Talk2026
PSU Threats and Cybersecurity Guest Lecture2025
UCSD Security and Cryptography Group2025
Cornell-Tech Trustworthy AI Guest Lecture2025
Google DeepMind Adversarial ML Seminar2025
Duke Generative AI Guest Lecture2025
UC Berkeley Security Seminar2024
Hong Kong Baptist University TMLR Young Scientist Seminar2024
Shanghai Jiao Tong University PAMI Group2024