Yihan Zhu
150K Fitzpatrick
South Bend, IN 46637
I’m a second-year PhD student from CSE, University of Notre Dame. I am very fortunate to be advised by Prof. Meng Jiang and working in DM2: Data Mining towards Decision Making lab.
I received my Bachelor’s degree in Electrical Engineering from Southeast University, China, in 2021. I was selected as the MS Honors Students at Columbia University in Spring 2022 and graduated with the Master of Science Award of Excellence in 2023.
My research focuses on AI for Science, 1) Graph representation learning, 2) Multimodal scientific reasoning.
My long-term goal is to build generative scientific world models that can simulate, explain, and reason about complex chemical systems. I aim to bridge scientific reasoning with reflective agent design, enabling models that not only generate hypotheses and molecular structures, but also critically evaluate, revise, and justify their own reasoning processes.
news
| Oct 11, 2025 | I am excited to receive the NeurIPS 25 travel award. See you in UCSD! |
|---|---|
| Sep 18, 2025 | The paper Learning Repetition-Invariant Representations for Polymer Informatics is accepted to NeurIPS! ✨ |
| Jul 21, 2025 | Our book 📘 Modeling Polymers with Neural Networks for polymer scientists interested in applying machine learning and neural networks is published by the American Chemical Society! |
| Jun 16, 2025 | The Open Polymer Challenge: Leveraging Machine Learning for Polymer Informatics was accepted to NeurIPS 2025 Competition Track and is now LAUNCHed on Kaggle  ! |
| Jun 06, 2025 | torch-molecule released! pip install torch-molecule to start developing your own models. |
selected publications
- NeurIPS 2025Learning Repetition-Invariant Representations for Polymer InformaticsarXiv preprint arXiv:2505.10726, 2025