Scholar
Isha Puri
Google Scholar ID: HTf7H58AAAAJ
PhD Student - AI/NLP@MIT, NSF Fellow, MIT Great Educators Fellow
Citations & Impact
Citations
301
H-index
5
i10-index
3
Publications
11
Co-authors
14
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'Beyond Binary Rewards: Training LMs to Reason About Their Uncertainty' accepted at COLM 2025 with Spotlight Oral presentation; cited by OpenAI’s latest work on hallucinations
- Paper 'MedPAIR: Measuring Physicians and AI Relevance Alignment in Medical Question Answering' presented at KDD Workshop on Trust & Evaluation in GenAI and COLM 2025
- Proposed 'Rollout Roulette', a particle filtering approach to scale inference in small open LMs without training, as a joint MIT CSAIL / Red Hat project
- Research on bias in LLMs for mental health support covered by MIT News and published in EMNLP Findings 2024
- Invited talks at NVIDIA, UT Austin, Stanford, University of Cambridge, Cohere AI, RedHat, and IBM Research AI
- Serving as Associate Program Chair for NeurIPS 2025
