About

• I am currently a Research Assistant at Wave Intelligence Lab, Department of Computer Science and Engineering, The Chinese University of Hong Kong, where I have been working with Prof. Shengchao Liu since June 2025.
• From July 2025 to Oct. 2025, I was a visiting student at Hefei Institutes of Physical Science, Chinese Academy of Sciences, hosted by Prof. Jianhui Zhou.
• I received my Bachelor of Science degree from the School of Physics, Sun Yat-sen University, in June 2025. During my undergraduate studies, I was a member of the PMI Lab and was advised by Prof. Haiping Huang from Oct. 2022 to June 2025.

Interests

My research interests lie at the intersection of artificial intelligence, statistical physics, and complex systems. I am particularly interested in understanding the theoretical mechanisms of large language models, generative models, and deep learning. Methodologically, I use tools from statistical mechanics, spin-glass theory, random matrix theory, dynamical mean-field theory, nonlinear dynamics. I am also developing a broad interest in AI for science.

News

2026-01-19	I formally joined the Wave Intelligence Lab at CUHK as a Research Assistant.
2025-06-20	I received my B.S. degree in Physics from Sun Yat-sen University.
2025-05-21	My undergraduate thesis was awarded the Excellent Undergraduate Thesis of Sun Yat-sen University.

Publications

* denotes equal contribution

• On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective
  Yuhao Li, Shengchao Liu
  Submitted to NeurIPS 2026. [arxiv]

• Base-and-Sugar Dual-Frame Flow Matching for RNA Co-Design
  Junzhe Li, Lijian Peng, Yuhao Li, Yize Zhou, Hanqun Cao, Cheng Tan, Shengchao Liu
  Submitted to NeurIPS 2026. [ICML 2026 Workshop GenBio]

• Spatiotemporal Perceptron with Positional Encoding: A State-Space Perspective
  Tianyi Shi, Hanqun Cao, Yuhao Li, Pheng-Ann Heng, Weitao Du, Shengchao Liu
  Submitted to NeurIPS 2026.

• A Minimal Model of Representation Collapse: Frustration, Stop-Gradient, and Dynamics
  Louie Hong Yao*, Yuhao Li*, Shengchao Liu
  Under review at Phys. Rev. E. [arxiv]

• Learning Mechanism and Phase Transitions of Triplet Loss for Contrastive Learning
  Yixiong Ren, Yuhao Li, Jianhui Zhou
  Under review at Phys. Rev. E.

• Spin Glass Model of In-Context Learning
  Yuhao Li, Ruoran Bai, and Haiping Huang
  Phys. Rev. E 112, L013301, 2025. [arXiv]