Training-method researcher

Stefano Ermon

ORCID lookup ↗DBLP lookup ↗Scholar lookup ↗Semantic Scholar ↗arXiv ↗

Co-created Direct Preference Optimization, aligning models from preference data without a separate reward model or RL loop.

Works in the archive (3)

FlashAttention (1/2/3): IO-Aware Exact Attention · 2022
Direct Preference Optimization (DPO) · 2023
RWKV / Hyena / Attention-SSM Hybrids (Jamba) · 2023

“Lead” marks the first-listed author — a byline fact, not a measure of influence; co-authorship is shown as recorded.

Co-authors in the archive

Christopher Ré (2), Tri Dao, Daniel Y. Fu, Atri Rudra, Jay Shah, Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher Manning, Chelsea Finn, Bo Peng, Michael Poli

View co-author network (15 people)

Loading graph…