Wenkai Li

I am a master studnet in LTI at Carnegie Mellon University, advise by Prof. Mona Diab, and work with Profs. Maarten Sap, Daniel Fried and Niloofar Mireshghallah. My research explores the AI safety, LLM Reasoning, Social Responsible AI, Interpretability, Agentic system and Human-AI Interaction. I received my Bachelor’s degree in Software Engineering from Northeastern University (China). During my undergraduate, I was interested in Software development and completed projects in distributed system and microservices system development.

My research goal focuses on building reliable and deployable AI agents capable of operating robustly in real-world environments. I am particularly interested in enhancing LLMs’ social and causal reasoning abilities to better align their behavior with human expectations, ultimately making LLM-based agents more reliable, safe, interpretable, and socially accepted.