Scholar
Mikhail Yurochkin
Google Scholar ID: QjBF9sUAAAAJ
Staff AI Scientist, IFM MBZUAI, ex MIT-IBM Watson AI Lab
Citations & Impact
Citations
4,982
H-index
26
i10-index
42
Publications
20
Co-authors
19
list available
Publications
Resume (English only)
Academic Achievements
- Several papers accepted at NeurIPS 2024, Findings of EMNLP 2024, COLM 2024, ICML 2024, ICLR 2024, and other top conferences. Involved in projects such as LLM routing, LoRA compression, and tinyBenchmarks.
Research Experience
- Currently a Staff AI Scientist at the IFM MBZUAI Silicon Valley Lab, leading data mixing for LLM pre-training. Previously, he was a research manager at the MIT-IBM Watson AI Lab, leading the Statistical Large Language Modeling group.
Education
- PhD in Statistics from the University of Michigan, advised by Prof. Long Nguyen.
Background
- Interested in a variety of LLM-related problems—pre- and post-training, data quality, reasoning, evaluation, routing, and efficient inference—and enjoy exploring statistical modeling approaches to solve them. Also worked on OOD generalization, algorithmic fairness, optimal transport, federated learning, and Bayesian nonparametrics.
Miscellany
- Continued to serve as Project Advisor for Break Through Tech AI in 2024.
