Scholar
Eric Wallace
Google Scholar ID: SgST3LkAAAAJ
UC Berkeley
Citations & Impact
Citations
22,466
H-index
42
i10-index
48
Publications
20
Co-authors
20
list available
Publications
Resume (English only)
Academic Achievements
- Published multiple papers on enhancing the security/privacy/robustness of ML, improving large language models, and their intersection. Notable works include 'Scalable Extraction of Data from (Production) Language Models' and 'The False Promise of Imitating Proprietary LLMs'.
Research Experience
- At OpenAI, works on safety and capabilities research, heavily involved in post-training and alignment for major models including GPT-4o mini, o1 and o1-preview, o1-mini, Instruction Hierarchy, o3 and o4-mini, and Deep Research.
Education
- PhD - University of California, Berkeley, Advisors: Dan Klein and Dawn Song
Background
- A researcher at OpenAI, working to make the next-generation of LLMs more safe, robust, and private. Previously, he did a PhD at UC Berkeley.
