Scholar
Jerry Wei
Google Scholar ID: Y4sk3aMAAAAJ
Anthropic
Citations & Impact
Citations
1,965
H-index
15
i10-index
15
Publications
18
Co-authors
9
list available
Publications
Resume (English only)
Academic Achievements
- Published papers: Constitutional classifiers: Defending against universal jailbreaks across thousands of hours of red teaming; Long-form factuality in large language models; Simple synthetic data reduces sycophancy in large language models; Symbol tuning improves in-context learning in language models; Larger language models do in-context learning differently.
Research Experience
- Lead the deployment-robustness team at Anthropic; was a research engineer at Google DeepMind and Google Brain; interned as a software-engineering intern at Meta in 2022.
Education
- Undergraduate at Stanford University, specializing in AI.
Background
- An AI researcher focusing on deployment robustness; based in San Francisco.
Miscellany
- Participated in several conference talks including Caltech AI Alignment, Generative AI Summit - Silicon Valley, Princeton AI Alignment, etc.
