Research / selected work

Evidence before biography.

My work moves from representing scientific systems, to learning their dynamics, to testing what modern AI systems actually understand.

Publications & preprints

07 records
  1. 2026
    MathAdv: What Theorem Provers Know, Reason, Formalize, and Generalize

    J. Yuan, C. M. Lockhart, X. Liu, J. Wang, C. Deng, X. Han, V. Mastrantonis, et al. · arXiv

    Reasoning↗
  2. 2025
    Omni-Mol: Multitask Molecular Model for Any-to-any Modalities

    C. Hu, H. Li, Y. Yuan, Z. Song, C. Zhao, H. Wang · NeurIPS

    Multimodal science↗
  3. 2025
    FMint: Bridging Human Designed and Data Pretrained Models for Differential Equation Foundation Model for Dynamical Simulation

    Z. Song, J. Yuan, H. Yang · Advanced Theory and Simulations

    World models↗
  4. 2025
    A Finite Expression Method for Solving High-Dimensional Committor Problems

    Z. Song, M. K. Cameron, H. Yang · SIAM Journal on Scientific Computing

    Scientific ML↗
  5. 2025
    Myelin Mapping in the Human Brain Using an Empirical Extension of the Ridge Regression Theorem

    G. S. Hampton, R. Neff, Z. Song, M. Bouhrara, R. Balan, R. G. Spencer · bioRxiv

    Medical imaging↗
  6. 2024
    Forecasting High-Dimensional Spatio-Temporal Systems from Sparse Measurements

    J. Song, Z. Song, P. Ren, N. B. Erichson, M. W. Mahoney, X. S. Li · Machine Learning: Science and Technology

    Reconstruction↗
  7. 2024
    Recent Advances on Machine Learning for Computational Fluid Dynamics: A Survey

    H. Wang, Y. Cao, Z. Huang, Y. Liu, P. Hu, X. Luo, Z. Song, et al. · arXiv

    Fluid dynamics↗