Pipi Hu - BIMSA

Office: A11-106

Email: hpp@bimsa.cn

Research Field: Large Foundation Model

Biography

Pipi Hu received his Bachelor's degree from Tsinghua University in 2014 and his Ph.D. in Mathematics from Tsinghua University in 2020. During his postdoctoral period from 2020 to 2022, he conducted interdisciplinary research in artificial intelligence and mathematics at Tsinghua University & BIMSA. In 2022, he joined Microsoft Research as a Senior Researcher, focusing on science foundation models. In 2025, he joined BIMSA as an Associate Professor. Pipi Hu's research interests primarily revolve around methods and applications in foundation models, including autoregressive diffusion models, discrete diffusion models, and science foundation models.

Research Interest

Research on cutting-edge technologies in foundation models, including continuous and discrete diffusion models as well as autoregressive models.
The scientific foundation model for science.
Machine learning based on differential equations: solving operators, inverse problems, and data-driven machine learning modeling with differential equation structures.

Education Experience

2014 - 2020 Tsinghua University Mathematics Ph.D
2010 - 2014 Tsinghua University Engineering Physics B.Sc.

Work Experience

2025 - BIMSA Associate Professor Large Foundation model
2022 - 2025 Microsoft Research Senior Research Fellow Foundation Model
2020 - 2022 BIMSA & YMSC Postdoc Deep Learning and Differential Equations

Publication

[1] Liwei Lu, Zhijun Zeng, Yan Jiang, Yi Zhu, and Pipi Hu, Weak Collocation Regression method: fast reveal hidden stochastic dynamics from high-dimensional aggregate data, Journal of Computational Physics, 502, 112799 (2024)
[2] L Guo, L Lu, Z Zeng, P Hu, Y Zhu, Weak Collocation Regression for Inferring Stochastic Dynamics with L\'{e} vy Noise, arXiv preprint arXiv:2403.08292 (2024)
[3] Zhijun Zeng, Pipi Hu, Chenglong Bao, Yi Zhu, Zuoqiang Shi, Reconstruction of dynamical systems from data without time labels, arXiv preprint arXiv:2312.04038 (2023)
[4] Guochang Lin, Pipi Hu, Fukai Chen, Xiang Chen, Junqing Chen, Jun Wang, Zuoqiang Shi, BINet: Learning to Solve Partial Differential Equations with Boundary Integral Networks, CSIAM Transactions on Applied Mathematics(2023)
[5] P. Hu, P. Xie, Yi Zhu, Traveling edge states in massive Dirac equations along slowly varying edges, IMA Journal of Applied Mathematics, 88(3), 455-471 (2023)
[6] Pipi Hu, Wuyue Yang, Yi Zhu, and Liu Hong, Revealing hidden dynamics from time-series data by ODENet, Journal of Computational Physics, 461, 111203 (2022)
[7] Pipi Hu, Ablowitz, Mark J.,Justin T.Cole,Pipi Hu,and Peter Rosenthal, Peierls-Nabarro barrier effect in nonlinear Floquet topological insulators, Phys. Rev. E, 103(4), -, 2021(2021)
[8] P Hu, L Hong, Y Zhu, Linear and nonlinear electromagnetic waves in modulated honeycomb media, Studies in Applied Mathematics, 144(1), 18-45 (2020)
[9] G Zhang, Y Li, R Luo, P Hu, Z Zhao, L Li, G Liu, Z Wang, R Bi, K Gao, ..., Unigenx: Unified generation of sequence and structure with autoregressive diffusion, arXiv (2025)
[10] M Ma, G Liu, C Cao, P Deng, T Dao, A Gu, P Jin, Z Yang, Y Xia, R Luo, ..., HybriDNA: A Hybrid Transformer-Mamba2 Long-Range DNA Language Model, arXiv (2025)
[11] L Guo, Z Wang, C Liu, J Li, P Hu, Y Zhu, Potential Score Matching: Debiasing Molecular Structure Sampling with Potential Energy Guidance, arXiv (2025)
[12] MR AI4Science, MA Quantum, The impact of large language models on scientific discovery: a preliminary study using gpt-4, arXiv (2023)
[13] G Lin, F Chen, P Hu, X Chen, J Chen, J Wang, Z Shi, Bi-greennet: learning green’s functions by boundary integral network, Communications in Mathematics and Statistics, 11(1), 103-129 (2023)
[14] W Cui, H Zhang, H Chu, P Hu, Y Li, On robustness of neural ODEs image classifiers, Information Sciences, 632, 576-593 (2023)
[15] G Lin, F Chen, P Hu, X Chen, J Chen, J Wang, Z Shi, A neural network framework for learning Green's function

Update Time: 2025-08-05 17:00:10