Hadi Khalaf
Ph.D. Student
Harvard University
I'm a first-year PhD student at Harvard SEAS. My research focuses on developing robust, trustworthy tools for LLM alignment.
Education
2024—Present
Harvard University
Ph.D. in Engineering Sciences
Advisor: Prof. Flavio Calmon
2020—2024
American University of Beirut
B.S. in Statistics and B.E. in Computer Engineering
Advisor: Prof. Ibrahim Abou Faycal
Experience
Summer 2023
Research Intern — Economics Department at Harvard
Advisor: Prof. Elie Tamer
Developed tools for counterfactual estimation in binary games.
Questions I'm Thinking About
Pairwise Preferences and Human Values
Can pairwise methods effectively capture subtle and diverse human values?
Reward Models as Auditing Tools
How can reward models serve as transparent and robust AI auditing mechanisms?
Scalable Oversight with Targeted Feedback
How can targeted human feedback guide LLMs towards better alignment?
Alignment in Multi-Agent Systems
How can we maintain aligned behavior among multiple AI agents with conflicting goals?