- Investigating the interaction of linguistic and mathematical reasoning in language models using multilingual number puzzles, EMNLP 2025
- Data Drives Unstable Hierarchical Generalization in LMs, EMNLP 2025
- To Backtrack or Not to Backtrack: When Sequential Search Limits Model Reasoning, COLM 2025
- What is the Right Notion of Distance between Predict-then-Optimize Tasks?, UAI 2025
- DDEQs: Distributional Deep Equilibrium Models through Wasserstein Gradient Flows, AISTATS 2025
- Mixture of Parrots: Experts improve memorization more than reasoning, ICLR 2025
- A Label is Worth a Thousand Images in Dataset Distillation, NeurIPS 2024
- Projects: OTDD has been incorporated into the DataSimilarity R package
Research Experience
- Worked at CSAIL, MIT, on various topics in machine learning and natural language processing
- Spent one year at IBM's T.J. Watson Research Center, working in the Speech Recognition and NLP teams
- Currently an Assistant Professor at Harvard SEAS, leading the Data-Centric Machine Learning (DCML) group
Education
- PhD: Massachusetts Institute of Technology (MIT), Computer Science
- Advisor: Not mentioned
- Time: Not mentioned
- Field: Machine learning and natural language processing
- MS: Courant Institute (NYU), Mathematics
- BSc: ITAM, Mathematics
Background
- Research Interests: Making machine learning more broadly applicable (especially to data-poor applications) and trustworthy (e.g., robust and interpretable)
- Field: Computer Science
- Bio: Assistant Professor at Harvard SEAS, leading the Data-Centric Machine Learning (DCML) group, also an Associate Faculty at the Kempner Institute, and affiliated with the Center for Research on Computation and Society and the Harvard Data Science Initiative. Also a researcher at Microsoft Research New England.