Scholar
Michihiro Yasunaga
Google Scholar ID: SieJYoEAAAAJ
Stanford University
Citations & Impact
Citations
23,951
H-index
41
i10-index
49
Publications
20
Co-authors
20
list available
Publications
Resume (English only)
Academic Achievements
- Published numerous papers covering topics such as multimodal reward benchmarks, robustness of reward models, content-adaptive image tokenization, large language models as analogical reasoners, and more. Some of these papers were presented at top conferences like ICLR 2024, NeurIPS 2024, etc.
Research Experience
- Worked as a researcher at Google DeepMind and Meta.
Education
- Received a PhD in Computer Science from Stanford, advised by Percy Liang, Jure Leskovec, and Chris Manning.
Background
- Interested in building LLMs and agents that assist humans in diverse tasks. His work spans: Post-training (RL, reward models, and evaluation), reasoning, retrieval and tool use for LLMs, and multimodality (vision-language models).
