CV

Education

• Ph.D. in Informatics, Pennsylvania State University, Aug 2024 – May 2028 (Expected)
  • Research focusing on LLM Security, AI agents and Evaluation
  • Advised by Prof. Jinyuan Jia
• B.Eng. in Computer Science, Wuhan University, Sep 2020 – Jun 2024

Experience

• Research Scientist Intern, ByteDance, May 2026 – Present
  • Working on Doubao Safety and Security
• Research Assistant, Pennsylvania State University, Aug 2024 – Present

Research Experience

My research focuses on developing secure and reliable LLMs and AI agents. I study both the vulnerabilities and defense / alignment of LLMs, with a particular focus on prompt injection. I also work on evaluating LLMs and AI agents under realistic and comprehensive settings.

• Prompt Injection Attacks and Defenses: I study how prompt injection affects LLM behavior and design efficient and effective defenses against these attacks.
• LLM Alignment and Post-Training: I have hands-on experience with reinforcement learning (RL) finetuning methods such as Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO). My research leverages post-training to further improve LLM robustness without degrading utility.
• Evaluation and Benchmarking: I build systematic evaluation pipelines to assess LLMs and AI agents. I am a co-author of a widely-used benchmark, Open-Prompt-Injection, and I lead PIArena, a next-generation evaluation platform for trustworthy LLMs and AI agents.

Publications

A full list is available on the homepage and on my Google Scholar profile.

Open Source Projects

• Open-Prompt-Injection – 480+ GitHub stars, co-author
• PoisonedRAG – 290+ GitHub stars, lead contributor and co-first author
• PIArena – lead contributor and first author
• PISanitizer – lead contributor and first author