CV

Research Intern

Shanghai Artificial Intelligence Laboratory

Working on Flow Matching-based chemical retrosynthesis generation algorithms

• Designed and implemented Flow Matching-based chemical retrosynthesis generation algorithms
• Participated in constructing end-to-end retrosynthesis prediction systems

Research Intern

Peking University & King's College London

Game-theoretic approaches for combinatorial optimization

• Applied reinforcement learning and game theory methods to combinatorial optimization
• ICLR 2022 Workshop Spotlight Presentation

Undergraduate Research Assistant

CAS Key Laboratory of Big Data Mining and Knowledge Management

Research on generative adversarial networks

• Participated in GAN-related theoretical research
• Outstanding Undergraduate Thesis: A Comprehensive Survey on Generative Adversarial Networks

PhD

The Chinese University of Hong Kong, Shenzhen

Computer Science

• Machine Learning
• Probabilistic Inference
• Diffusion Models

Master of Science

University of Chinese Academy of Sciences

Operations Research and Cybernetics

Bachelor of Science

Zhengzhou University

Mathematics and Applied Mathematics

Shenzhen Institute of Big Data PhD Student Scholarship

Shenzhen Institute of Big Data

ICLR 2022 Workshop Spotlight Presentation

ICLR Workshop on Gamification and Multiagent Solutions

Outstanding Participant in Oxford International Exchange Program

Oxford University

Outstanding Undergraduate Thesis Award

Zhengzhou University

A Comprehensive Survey on Generative Adversarial Networks

Efficient Training of Multi-task Neural Solver for Combinatorial Optimization

TMLR 2025

Designed Multi-armed Bandits-based task selection strategies for multi-task optimization.

Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory

KDD 2025

Analyzed oversmoothing phenomena in diffusion-based continuous GNNs. (CCF A-class)

Sampling from Binary Quadratic Distributions via Stochastic Localization

ICML 2025

Proposed efficient discrete sampling algorithms based on stochastic localization. (CCF A-class)

ASP: Learn a Universal Neural Solver!

IEEE TPAMI 2024

Proposed ASP universal neural routing optimization framework. (CCF A-class)

Learning Graph Representation by Aggregating Subgraphs via Mutual Information Maximization

Neurocomputing 2023

Constructed graph representation learning framework based on mutual information maximization.

Solving uncapacitated P-Median problem with reinforcement learning assisted by graph attention networks

Applied Intelligence 2023

Diffusion Process-based Sampling Methods

Proposed efficient discrete sampling algorithms and Neural Samplers based on Diffusion Models

• ICML 2025 paper on stochastic localization
• Importance Sampling-based diffusion samplers
• High-performance PyTorch/JAX implementation

Combinatorial Optimization and Neural Solvers

ASP universal neural routing optimization framework with game-theoretic self-play mechanisms

• IEEE TPAMI 2024 publication
• Multi-task learning with Bandit algorithms
• Deep reinforcement learning methods

Graph-based Deep Learning

Theoretical analysis and practical applications of graph neural networks

• KDD 2025 paper on diffusion-based GNNs
• Operator semigroup theory perspective
• Graph representation learning