• 2008–2012: Jilin University, Computer Science and Technology, Bachelor’s Degree (Full-time)
• 2012–2015: Jilin University, Computer Science and Technology, Master’s Degree (Full-time)
• 2015–2021: Jilin University, Computer Science and Technology, Doctoral Degree (Full-time)
• 2016–2017: Hong Kong Baptist University, Department of Computer Science, Research Assistant
• 2019–2020: University of Illinois at Urbana–Champaign, Department of Computer Science, Visiting scholar

Work Experience

• 2021–2025: Faculty of Electronic and Information Engineering, Xi’an Jiaotong University, Assistant Professor
• 2025–Present: Faculty of Electronic and Information Engineering, Xi’an Jiaotong University, Associate Professor
• 2025–Present: Center for Intelligent Science and Engineering, Shenzhen Loop Area Institute, Joint-appointed Professor

Hongbin Pei, Ph.D., is an Associate Professor at Xi’an Jiaotong University and a jointly appointed Professor at Shenzhen Loop Area Institute. He received his B.S., M.S., and Ph.D. degrees from Jilin University, and was a visiting scholar at the University of Illinois at Urbana–Champaign. His research interests include graph learning, trustworthy reasoning in large models, and AI-driven molecular science. He has published more than 40 papers in CCF-A venues such as IEEE TPAMI and ICML. His proposed Geometric Graph Convolutional Network (Geom-GCN) has been recognized as a foundational work in heterogeneous graph learning. He has led multiple national research projects,  including the NSFC General Program and Young Scientists Fund. His work has been applied to large-scale industrial, urban planning, and public health systems. His science and technology policy recommendations have been adopted by central government offices.

Representative Publications (past 5-10 years, in order of impact)：

1. Hongbin Pei, et al.  Active Surveillance via Group Sparse Bayesian Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022（IEEE TPAMI）
2. Hongbin Pei, et al. Geom-GCN: Geometric Graph Convolutional Networks. The eighth International Conference on Learning Representations, 2020（ICLR; Spotlight Paper; Citation 1800+）
3. Hongbin Pei, et al. Multi-Track Message Passing: Tackling Oversmoothing and Oversquashing in Graph Learning via Preventing Heterophily Mixing. The forty-first International Conference on Machine Learning, 2024（ICML; Spotlight Paper）  
4. Hongbin Pei, et al. Non-Stationary Predictions May Be More Informative: Exploring Pseudo-Labels with a Two-Phase Pattern of Training Dynamics. The forty-first International Conference on Machine Learning, 2025（ICML）
5. Huiqi Deng, Hongbin Pei, et al. Attribution Explanations for Deep Neural Networks: A Theoretical Perspective. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026（IEEE TPAMI）
6. Hongbin Pei, et al. Challenges, countermeasures, and forward-looking technologies for cyber society governance from system security perspective. Bulletin of Chinese Academy of Sciences, 2025

Books/Monographs：

• Strategic Research on Information Technology Supporting the Modernization of National Governance, Science Press, 2025, Co-author.
• Introduction to Big Data and Artificial Intelligence, Posts & Telecom Press, 2023, Co-author.
•  

Patents：

• A Pseudo-Label–Enhanced Training Method Using Prediction-Inconsistent Samples, ZL202510525771.0, 2025, First Inventor.
• An Adversarial Training Method for Deep Force Field Models for Stable Simulation, ZL202411330354.2, 2024, First Inventor.
• A Method for Determining Forces in Perturbed Molecular Conformations**, ZL202411328773.2, 2024, First Inventor.

Research Awards：

• 2021: First Prize of Jilin Provincial Natural Science Award, People’s Government of Jilin Province, Fourth Completed Person.
• 2023: Wu Wenjun Artificial Intelligence Excellent Doctoral Dissertation Nomination Award, Chinese Association for Artificial Intelligence, First Completed Person.