Google Scholar
Education
Ph.D. in Computer Science Northwestern University
B.S. in Physics National Taiwan University
About
Jerry Yao-Chieh Hu is an Assistant Professor of Computer Science at Vanderbilt University in the College of Connected Computing (CCC). His research is in Machine Learning and Artificial Intelligence, with a focus on foundation models.
Recently, he explores the principles underlying how foundation models learn, store knowledge, compute, adapt, and generalize. He uses these principles to develop new models and learning methods and to study applications in science.
Jerry received his Ph.D. in Computer Science from Northwestern University, advised by Prof. Han Liu, and his B.S. in Physics from National Taiwan University, advised by Prof. Pisin Chen.
Research Focus
His research focuses on the science and technology of foundation models and related AI systems (including LLMs, generative models, world models, and agentic systems) through three connected aims:
- Foundations: Theory for how foundation models learn, store knowledge, compute, adapt, and generalize.
- Methods: Principled architectures and algorithms for large foundation models.
- Science: Foundation models for scientific discovery (e.g., genomics, virtual cells, drug design, physics).
Publication Highlights
In-Context Universal Approximation, Compositional Generalization, and Algorithm Emulation. Jerry Yao-Chieh Hu, Hong-Yu Chen, Po-Chiao Lin, Maojiang Su, and Han Liu. International Conference on Machine Learning (ICML), 2026.
On Sparse Modern Hopfield Model. Jerry Yao-Chieh Hu, Donglin Yang, Dennis Wu, Chenwei Xu, Bo-Yu Chen, and Han Liu. Conference on Neural Information Processing Systems (NeurIPS), 2023.