Hi, I’m Raj! I’m a CS PhD student at Cornell Tech, advised by Emma Pierson. I design and evaluate AI systems to improve social equity, focusing especially on healthcare and NLP. My healthcare work focuses on evaluating fairness (MLHC 2023) and designing equity-focused ML methods (arXiv 2023). My NLP work focuses on pluralistic alignment: where do users disagree about what constitutes a “good” language model, and how do we engage with disagreement in model design (EMNLP 2024)?
In 2024, I interned with the FATE group at Microsoft Research Montréal. At Cornell, I am part of the Digital Life Initiative and the AI, Policy, and Practice working groups. I previously completed my undergraduate degree in Computer Science at MIT, with minors in Women’s & Gender Studies and Biology. I enjoyed working with the Data + Feminism Lab.
I occasionally write things here. I’m passionate about food (don’t tell my advisor that I sometimes think more about cooking than research), and I occasionally write food-related things on Substack. Finally, it takes a village to raise a PhD student; I am deeply indebted to mine. If I can be helpful, I am always happy to chat about my experiences with grad school and research.
Recently:
- [Sep 2024] Our work on LLM annotation alignment is accepted at EMNLP 2024!
- [Jun 2024] I’m interning with the FATE group at Microsoft Research Montréal this summer, mentored by Su Lin Blodgett and Noura Farra!
- [Jun 2024] I’ll be at NAACL in Mexico City! I’m presenting a survey on shifts in LLM publication trends in 2023 in the main conference, and on how LLMs evaluate safety compared to human annotators at the Workshop for Online Abuse and Harms.
- [Dec 2023] Our paper on granular race disparities won a Best Paper Honorable Mention at ML4H 2023!
Selected Work
Annotation alignment: Comparing LLM and human annotations of conversational safety.
Rajiv Movva, Pang Wei Koh, Emma Pierson.
EMNLP 2024.
[paper] [twitter]
Topics, Authors, and Institutions in Large Language Model Research: Trends from 17K arXiv Papers.
Rajiv Movva*, Sidhika Balachandar*, Kenny Peng*, Gabriel Agostini*, Nikhil Garg, Emma Pierson.
NAACL 2024.
[paper] [twitter]
Coarse race data conceals disparities in clinical risk score performance.
Rajiv Movva*, Divya Shanmugam*, Kaihua Hou, Priya Pathak, John Guttag, Nikhil Garg, Emma Pierson.
MLHC 2023 (Proceedings) & ML4H 2023 (Findings).
🏆 Honorable Mention, Best Findings Paper 🏆, ML4H 2023.
[paper] [twitter] [Cornell news]
Towards Intersectional, Feminist, Participatory ML: A Case Study in Supporting Feminicide Counterdata Collection.
Harini Suresh, Rajiv Movva, Amelia Dogan, Rahul Bhargava, Isadora Cruxên, Ángeles Martinez Cuba, Giulia Taurino, Wonyoung So, Catherine D’Ignazio.
FAccT 2022.
🏆 Best Student Paper 🏆
[paper] [twitter]