Scholar
Sabri Eyuboglu
Google Scholar ID: ya1egC8AAAAJ
PhD Student in Computer Science, Stanford University
Citations & Impact
Citations
985
H-index
11
i10-index
12
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Published several papers, including 'Cartridges: Lightweight and general-purpose languge model memory via self-study' and 'Minions: Cost-efficient collaboration between on-device and cloud language models', and presented at ICML conferences.
Research Experience
- Currently researching at the Stanford Machine Learning Group, focusing on enhancing the ability of language models to store and recall large amounts of user-provided information. In the early days of his PhD, he developed techniques and tools to help deep learning practitioners identify subtle failure modes of their models. Worked as a machine learning researcher at Flatiron Health for a year.
Education
- PhD in Computer Science from Stanford University (advised by Chris Ré and James Zou); B.S. + M.S. in Computer Science from Stanford University (graduated in 2019)
Background
- PhD candidate in Computer Science at Stanford University, with research interests in improving the memory capabilities of large language models, particularly focusing on the tradeoff between cost and quality. Previously a machine learning researcher at Flatiron Health.
Miscellany
- Serves as an advisor to Cartesia, a startup applying some of his research to the problem of delivering real-time AI.
Co-authors