picture
  • Home
  • Research
  • News
  • Publications
  • Awards
  • Gallery
  • Experience
  • Zezhong Ding (丁泽中)

    My heart is in the research!

    ✉️ Email: zezhongding AT mail DOT ustc DOT edu DOT cn
    🏫 University of Science and Technology of China (USTC)

    Google Scholar, Github

    picture picture

    $ whoami

    Zezhong (aka Zeddy) is a Ph.D. candidate (successive master-doctor program since 2022) in Data Darkness Lab (DDL), School of Artificial Intelligence and Data Science, University of Science and Technology of China (USTC), supervised by Prof. Xike Xie. He was a visiting student at Department of Computer Science, Hong Kong Baptist University, from 2024 to 2025, working with Prof. Jianliang Xu. He received his B.Eng. (Hons) degree in Computer Science and Technology from Ocean University of China in 2022. He has published several research papers in prestigious conferences and journals such as SIGMOD, NeurIPS, ACL, and TC. His general research interests mainly focus on the data management, machine learning, and computer systems over big data and large models, particularly big graph data and large language models.

    $ echo "📌Welcome to contact me for any discussion!😊"


    $ head research_interests

    • 🎯Graph Partitioning: How to Partition Big Graphs?
      • [S5P, SIGMOD-24]: (1) Streaming Skewness-aware Graph Partitioner based on Stackelberg Game; (2) Play-and-Plug Integration with the Downstream Distributed Graph Computing System (e.g., PowerGraph)
      • [ClusPar, TC-25]: (1) Streaming Edge Clustering; (2) Play-and-Plug Integration with the Downstream Distributed Graph Computing System (e.g., PowerGraph)
    • 🎯Graph Machine Learning: How to Design Graph Machine Learning Models?
      • [DuetGraph, NeurIPS-25]: (1) Knowledge Graph Reasoning Model with Attention and Message Passing Modules; (2) Addressing the Over-Smoothing Challenge
      • [SamGoG, arXiv-25]: (1) Sampled Graph-of-Graphs for Graph Classification; (2) Addressing the Imbalance Challenge
    • 🎯LLMs: How to Enhance LLMs' Reasoning Ability?
      • [GraphInsight, ACL-25]: (1) Importance-based Strategy for Regions with Stronger Memory Performance; (2) RAG for Regions with Weaker Memory Performance
      • [GraphVista, arXiv-25]: (1) Multimodal Graph Understanding; (2) Reinforcement Learning for Chain-of-Thought Generation
    • 🎯Graph Training Systems: How to Train Big Graphs?
      • [Capsule, SIGMOD-25]: (1) Out-of-Core Graph Training Mechanism; (2) Play-and-Plug Implementation for DGL and PyG
      • [SWIFT, SIGMOD-26]: (1) Out-of-Core Dynamic Graph Training System; (2) Storage Optimization

    $ tail news

    • 2025/9/18One paper is accepted by NeurIPS 2025 (San Diego Convention Center & Mexico City)!
    • 2025/6/26I give an oral presentation at SIGMOD 2025 in Berlin, Germany!
    • 2025/5/23One paper is accepted by SIGMOD 2026 (Bengaluru, India)!
    • 2025/5/16One paper is accepted by ACL Main Conference 2025 (Vienna, Austria)!
    • 2025/2/10I have completed my three-month visit at HKBU DB Group! Best wishes for the HKBU DB Group!
    • 2024/11/1One paper is accepted by SIGMOD 2025! Next station is Berlin, Germany!
    • 2024/9/30One paper is accepted by IEEE Transactions on Computers (My first journal paper)!
    • 2024/6/13I give my first conference talk on SIGMOD 2024 in Santiago, Chile!
    • 2024/3/30Data Darkness Lab is recruiting graduate students, USTC! Inquiries Welcome!
    • 2024/2/23One paper is accepted by SIGMOD 2024 (My first conference paper)! Looking forward to the conference in Santiago, Chile!

    $ head publications

    You can find the full list of my publications here, my DBLP entry here and my Google Scholar entry here.

    $ head awards


    $ cat academic_services


    $ cat teaching_experiences


    $ cat misc