Safety researcher
Ethan Perez
Built red-teaming and evaluation methods for finding model failures and studied reward overoptimization in alignment.
Works in the archive (3)
- Retrieval-Augmented Generation (RAG) · 2020
- Reward Model Overoptimization / Sycophancy · 2022
- Reasoning Correctives: Faithfulness / Overthinking / Contamination · 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
View co-author network (10 people)
Loading graph…