I am a Ph.D. student in Computer Science at Shanghai Jiao Tong University (SJTU), supervised by Prof. Junchi Yan (严骏驰), and a member of the prestigious Wen-Tsun Wu AI Honorary Doctoral Class. I completed my undergraduate degree at SJTU in 2023, majoring in Artificial Intelligence within the Pilot Class for Outstanding Talent, an elite program reserved for top 5% of students.
My research centers on deep learning for complex optimization problems, with a particular emphasis on discrete combinatorial optimization. I have authored seven papers in leading conferences and journals, including four as (co-)first author at CVPR 2023, ICML 2024, NeurIPS 2025, and ICLR 2026. Additionally, I serve as a reviewer for top-tier venues, e.g., ICML, NeurIPS, ICLR, and TPAMI.
🔥 News
- 2026.01: 🎉 One paper on representation learning for combinatorial optimization is accepted by ICLR 2026!
- 2025.09: 🎉 One paper on sampling approach for combinatorial optimization is accepted by NeurIPS 2025!
📝 Publications
Combinatorial Optimization
ConRep4CO: Contrastive Representation Learning of Combinatorial Optimization Instances across Types
Ziao Guo, Yang Li, Shiyue Wang, Junchi Yan
- This work proposes a contrastive pre-training framework for combinatorial optimization (CO) through an instance-level contrastive learning objective that aligns original CO problem instances with their canonical SAT-form counterparts.
Fractional Langevin Dynamics for Combinatorial Optimization via Polynomial-Time Escape
Ziao Guo*, Shiyue Wang*, Changhong Lu, Junchi Yan
- This work proposes a sampling technique for solving combinatorial optimization problems, and theoretically proves that this approach exhibits a polynomial time for escaping local minima, compared with exponential escape time of original LD in previous work.
Ziao Guo, Yang Li, Chang Liu, Wenli Ouyang, Junchi Yan
- This work proposes a hardness-preserving MILP instance generation framework with adaptive constraint modification and constraint interrelation modeling.
Deep Learning of Partial Graph Matching via Differentiable Top-K
Ziao Guo*, Runzhong Wang*, Shaofei Jiang, Xiaokang Yang, Junchi Yan
- This work proposes the first neural solver for partial graph matching (GM) problem, and release a new benchmark for partial GM.
-
Pygmtools: A python graph matching toolkit
Runzhong Wang, Ziao Guo, Wenzheng Pan, Jiale Ma, Yikai Zhang, Nan Yang, et al. -
Towards Imitation Learning to Branch for MIP: A Hybrid Reinforcement Learning Based Sample Augmentation Approach
Changwen Zhang, Wenli Ouyang, Hao Yuan, Liming Gong, Yong Sun, Ziao Guo, et al. -
LinSATNet: the positive linear satisfiability neural networks
Runzhong Wang, Yunhao Zhang, Ziao Guo, Tianyi Chen, Xiaokang Yang, Junchi Yan.
🎖 Honors and Awards
- 2024 2025 Wen-Tsun Wu Scholarship
- 2024 Yang Jiachi Scholarship (Top 1%)
- 2023 Shanghai Jiao Tong University Merit Graduate
- 2021 Tencent Scholarship (Top 1%)
- 2021 Finalist Winner of Mathematical Contest in Modeling (Top 2%)
📖 Educations
- 2023.09 - present, PhD of Comupter Science and Technology, Shanghai Jiao Tong University (SJTU, 上海交通大学)
- 2019.09 - 2023.06, B.S in Artificial Intelligence (The Pilot Class for Outstanding Talent), Shanghai Jiao Tong University (SJTU, 上海交通大学)