Zexue He

I recently recieved my Ph.D. in Computer Science and Engineering department at University of California, San Diego, advised by Prof. Julian McAuley. I am now a researcher at the MIT-IBM Lab and an affiliate researcher at MIT CSAIL. Previously, I was an intern at the Microsoft Research and Google.

My research focuses on advancing Trustworthy NLP to address high-stakes human-centric tasks, e.g., in human healthcare and college admission. My works utilize human-centered design, diagnose the shortcomings of existing models, and develop AI-driven solutions that are more responsible and better satisifies human needs in real life, with less labor effort and cost:

• Trustworthy NLP: enhance the fairness, harmlessness, robustness, interpretability, and controllability of LLMs;
• AI for Social Good : build NLP for Healthcare, AI for Low-Resource, and AI for Therapy.
• Learning from Humans: understand human cognitive behaviors, brain functions, and reading behaviors

• Jan. 27th, 2025: I will give a talk at Tufts. Feel free to reach out if you want to chat!
• Jan. 6th, 2025: Invited talk on Responsible AI for High-Stakes Human-Centric Domains at UCLA. Thanks UCLA NLP group!
• Nov. 7th, 2025 : Invited talk on Responsible AI for Human-centered Challenges at MIT. Thanks MIT SiA-Lab!
• Oct. 13th, 2024: Talk on Trustworthy GenAI at UCSD CSE.