Yizhang (Jerry) He

Yizhang (Jerry) He

Associate Professor at Global College, Shanghai Jiao Tong University

About me

Yizhang He is currently an associate professor at the Global College, Shanghai Jiao Tong University. Before joining SJTU, he was an Associate Lecturer (Teaching & Research) in the School of Computer Science and Engineering at University of New South Wales from 2025 to 2026, and a Postdoctoral Fellow at UNSW in 2025. He was a member of the Data and Knowledge Research Group at UNSW. He got his Ph.D. in Computer Science from UNSW in 2025, his M.Phil. in Computer Science from UNSW in 2021, and his bachelor’s degree in Data Science from UNSW in 2019.

His research interests lie in big data analytics, large language models, and privacy-preserving AI. In particular, he works on graph mining, cohesive subgraph discovery, motif counting, and differential privacy for large-scale data systems.

Prospective Students

I am looking for self-motivated students for Ph.D., master’s, and research internship opportunities. Students interested in big data analytics, graph mining, privacy-preserving AI, and large language models are welcome to contact me.

Outstanding students may be recommended for opportunities at leading securities firms such as GF Securities and major technology companies such as Alibaba and ByteDance.

Research Directions

My current work spans the following directions. Detailed results are under submission or in preparation.

Graph Mining under Differential Privacy

Developing scalable algorithms for subgraph counting, densest subgraph discovery, and graph queries under local differential privacy. This line of work builds on my PhD thesis and continues with active collaborators.

Large Language Models for Data Management

Exploring how LLMs can be applied to classic database problems, including SQL query rewriting, relational data integration, and privacy-preserving question answering. These projects bridge my database background with the capabilities of modern language models.

LLM Security

Investigating both offensive and defensive aspects of LLM safety. On the attack side, developing black-box jailbreak methods that leverage structured attack memory and adaptive search strategies to systematically evaluate model vulnerabilities. On the defense side, exploring mechanisms to harden LLMs against adversarial prompts and unauthorized information disclosure. This line of work aims to bridge the gap between red-teaming practice and principled safety engineering.

LLM Agent Systems

Developing context management strategies for LLM-based autonomous agents to operate effectively over long-horizon, multi-step tasks. This includes learning when and how to compress, retain, or discard information across extended interactions, and designing lightweight adaptation methods that enable a context manager trained on one task to transfer to new environments with minimal re-training.

Publications

Also available on Google Scholar.

  1. Yizhang He, Wenjie Zhang, Kai Wang, Xuemin Lin, Ying Zhang, Wei Ni. Efficient and Effective Biclique Counting with Local Differential Privacy, ACM SIGMOD International Conference on Management of Data (SIGMOD) , 2026.
  2. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang, Wei Ni. Robust Privacy-Preserving Triangle Counting under Edge Local Differential Privacy, ACM SIGMOD International Conference on Management of Data (SIGMOD) , 2025.
  3. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang. Common Neighborhood Estimation over Bipartite Graphs under Local Differential Privacy, ACM SIGMOD International Conference on Management of Data (SIGMOD) , 2025.
  4. Yuting Zhang, Wei Ni, Kai Wang, Yizhang He, Conggai Li. Truss Decomposition Under Edge Local Differential Privacy, IEEE International Conference on Data Engineering (ICDE) , pages 2670-2683, 2025.
  5. Shunyang Li, Kai Wang, Wenjie Zhang, Xuemin Lin, Yizhang He. Efficient Bitruss Decomposition on GPU, IEEE Transactions on Knowledge and Data Engineering (TKDE) , 37(8): 4578-4590, 2025.
  6. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang. Discovering Critical Vertices for Reinforcement of Large-scale Bipartite Networks, The VLDB Journal (VLDBJ) , 33: 1861-1886, 2024.
  7. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Wei Ni, Ying Zhang. Butterfly Counting over Bipartite Graphs with Local Differential Privacy, IEEE International Conference on Data Engineering (ICDE) , pages 2351-2364, 2024.
  8. Shunyang Li, Kai Wang, Xuemin Lin, Wenjie Zhang, Yizhang He, Long Yuan. Querying Historical Cohesive Subgraphs over Temporal Bipartite Graphs, IEEE International Conference on Data Engineering (ICDE) , pages 2503-2516, 2024.
  9. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang. Scaling Up k-Clique Densest Subgraph Detection, ACM SIGMOD International Conference on Management of Data (SIGMOD) , 1(1): 1-26, 2023.
  10. Kai Wang, Gengda Zhao, Wenjie Zhang, Xuemin Lin, Ying Zhang, Yizhang He, Chunxiao Li. Cohesive Subgraph Discovery over Uncertain Bipartite Graphs, IEEE Transactions on Knowledge and Data Engineering (TKDE) , 35(11): 11165-11179, 2023.
  11. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang. Efficient Reinforcement of Bipartite Networks at Billion Scale, IEEE International Conference on Data Engineering (ICDE) , pages 446-458, 2022.
  12. Gengda Zhao, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang, Yizhang He. Efficient Computation of Cohesive Subgraphs in Uncertain Bipartite Graphs, IEEE International Conference on Data Engineering (ICDE) , pages 2333-2345, 2022.
  13. Yizhang He, Kai Wang, Wenjie Zhang, Xuemin Lin, Ying Zhang. Exploring Cohesive Subgraphs with Vertex Engagement and Tie Strength in Bipartite Graphs, Information Sciences , 572: 277-296, 2021.

Teaching

As Lecturer

  • COMP3311 Database Systems — 2025 T3, 2026 T1
  • ZZEN9311 Database Systems — 2025 T3

As Teaching Assistant (UNSW, 2020–2024)

  • Database Systems (COMP9311) — 2020 T3; 2021 T1, T2; 2023 T2, T3
  • Intro to Data Science (DATA1001) — 2020 T2; 2021 T2
  • Graph Analytics (COMP9312) — 2021 T2; 2024 T2
  • Big Data Management (ZZEN9313 / COMP9313) — 2021 T3

Academic Service

  • Reviewer: ICDE, SIGMOD, VLDBJ, TKDE

Selected Awards

  • Early Career Academic Fellowship (ECAF), UNSW, Aug 2025
  • SIGMOD Travel Award, ACM SIGMOD, 2025
  • Chinese Government Award for Outstanding Self-Financed Students Abroad (Category A), Jun 2025
  • HDR Development and Research Training Grant (DRTG), Oct 2023 & Apr 2024
  • Top-up Scholarship for Outstanding Research Performance, UNSW, Nov 2022
  • University International Postgraduate Award (UIPA), UNSW, Jun 2021