Scholar
Mehul Damani
Google Scholar ID: PBAmmGMAAAAJ
MIT
Citations & Impact
Citations
1,245
H-index
9
i10-index
9
Publications
15
Co-authors
1
list available
Publications
Resume (English only)
Academic Achievements
- 1. Trained reasoning models to reason about their uncertainty using RL, new paper published!
- 2. Started internship at MIT-IBM Watson Lab to work on RL for tool-use.
- 3. Paper on test-time training was accepted to ICML!
Research Experience
- 1. Worked with Lerrel Pinto at NYU on developing automatic curriculum learning methods for RL agents.
- 2. Part of the MARMot Lab at NUS, worked with Guillaume Sartoretti on applying multi-agent reinforcement learning to traffic signal control and multi-agent pathfinding.
Education
- Third year Ph.D. student at MIT, advised by Jacob Andreas.
Background
- Research interests lie at the intersection of reinforcement learning (RL) and large language models (LLMs). Focused on using RL to improve reasoning, math, coding, and other capabilities in LLMs, reducing hallucinations in LLMs, and thinking about how optimally selecting inference-time techniques can significantly improve the efficiency of LLMs.
Miscellany
- Always excited to explore new research directions and open to collaborating or advising students. If you are interested in my research or simply want to chat, don't hesitate to get in touch!
