Paper: 'The Ideation-Execution Gap: Execution Outcomes of LLM-Generated versus Human Research Ideas' (preprint), with Tatsunori Hashimoto and Diyi Yang; covered by Science.
Paper: 'Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers', ICLR 2025, with Diyi Yang and Tatsunori Hashimoto; featured in Nature News, Newsweek, Synced, etc.
Paper: 'Predicting Empirical AI Research Outcomes with Language Models', NeurIPS 2025, with He He et al.
Invited talks at Stanford, MIT, CMU, Intel, USC, Northeastern, MILA, Samaya AI, UVA, alphaXiv, Ai2, etc., on automating research.
Co-organizer of COLM 2025 in Montreal (LM4Sci workshop).