Zhengpin Li's Homepage

No.5 Yiheyuan Road
Haidian District
Beijing, China
Background. Zhengpin Li is a Postdoctoral Fellow at the Center of Machine Learning Research, Peking University, co-supervised by Wentao Zhang and Weinan E. He obtained his Ph.D. in Statistics from the School of Data Science, Fudan University in 2025, co-supervised by Yanxi Hou and Jian Wang. He received his B.Sc. in Mathematics from the School of Mathematics, Shandong University in 2019. He worked with Rex (Zhitao) Ying as a visiting student at the Department of Computer Science, Yale University in 2024.
Research Interests. Zhengpin’s research is primarily focused on graph mining, trustworthy machine learning, large language model, and extreme value theory. He has published several papers in top venues such as ICML, WebConf, and Neural Networks. Most recently, he is interested in the following topics,
- LLM reasoning based on knowledge graphs
- data-centric LLM pre- and post-training
- graph foundation models
Email: zpli [at] pku [dot] edu [dot] cn.
selected publications
- ICML 2025Retrieval-Augmented Language Model for Knowledge-aware Protein EncodingForty-Second International Conference on Machine Learning, 2025
- WebConf 2025Fairness-aware Prompt Tuning for Graph Neural NetworksProceedings of the ACM on Web Conference, 2025
- ICASSP 2024SAM: A Self-Adaptive Attention Module for Context-Aware Recommendation SystemIEEE International Conference on Acoustics, Speech and Signal Processing, 2024
- Neural Netw.Beyond Smoothness: A General Optimization Framework for Graph Neural Networks with Negative Laplacian RegularizationNeural Networks, 2024
- ICASSP 2024Spectral Graph Neural Networks with Generalized Laguerre ApproximationIEEE International Conference on Acoustics, Speech and Signal Processing, 2024
- Comput. Stat.Recent Advances on Mechanisms of Network Generation: Community, Exchangeability, and Scale-Free PropertiesWiley Interdisciplinary Reviews: Computational Statistics, 2024
- Stat. Comput.Online prediction of extreme conditional quantiles via B-spline interpolationStatistics and Computing, 2024
- ICASSP 2022Applying Differential Privacy to Tensor CompletionIEEE International Conference on Acoustics, Speech and Signal Processing, 2022