Zhuohang Li

I am a PhD candidate in Computer Science at Vanderbilt University, with research internship experience at Google, Amazon, MERL, and NEC Labs. I hold an MS from Rutgers and a BS from UESTC.

My research is in the field of trustworthy machine learning, where I work to make AI systems safer and more dependable. My work identifies and mitigates critical failure modes in modern machine learning models, with a focus on:

• Reliability in LLMs: Improving model dependability by assessing knowledge relevance, preventing hallucinations, ensuring factuality, and improving alignment with human preferences.
• Security and Privacy in AI: Defending against adversarial attacks and analyzing privacy risks in machine learning systems.