Yuan Li

Yuan Li

PhD Candidate@ZJU

Hi, I’m Yuan Li, a PhD student in the ZJU Programming Languages and Automated Reasoning Group at Zhejiang University, advised by Prof. Peisen Yao. Currently, I am a visiting PhD student in the Software Systems Engineering Group at University College London (UCL), supervised by Prof. He Ye.

My research interests lie in the intersection of programming languages, software engineering, cybersecurity, and artificial intelligence, with a particular focus on topics about the evaluation and optimization of techniques related to static analysis.

🔥 News

  • 2026.09.30:  🎉 DatalogBench released – a benchmark for text-to-Datalog synthesis.
  • 2025.04.02:  🎉 Our work on SAST Evaluation has been accepted by FSE 2025.
  • 2024.10.11:  🎉 I passed the Mid-term Assessment of my PhD dissertation research.

📝 Publications

  • [Arxiv’26] DatalogBench: Evaluating Large Language Models on Text-to-Datalog Synthesis
    Yuan Li, Hanyun Jiang, Guowei Tian, Chengpeng Wang, Peisen Yao
    🔗 [Paper] 💻 [Code]

  • [FSE’25] Understanding Industry Perspectives of Static Application Security Testing (SAST) Evaluation
    Yuan Li, Peisen Yao, Kan Yu, Chengpeng Wang, Yaoyang Ye, Song Li, Meng Luo, Yepang Liu, Kui Ren
    The ACM International Conference on the Foundations of Software Engineering (CORE-A*/CCF-A)
    🔗 [Paper] 📖 [Codebook]

  • [TITS’23] Imbalanced Malware Family Classification Using Multimodal Fusion and Weight Self-Learning
    Shudong Li, Yuan Li (Co-1st & 1st Student), Xiaobo Wu, Sattam Al Otaibi, Zhihong Tian
    IEEE Transactions on Intelligent Transportation Systems (JCR-Q1/CCF-B)
    🔗 [Paper] 💻 [Code]

  • [SMPT’21] Malicious Mining Code Detection Based on Ensemble Learning in Cloud Computing Environment
    Shudong Li, Yuan Li (Co-1st & 1st Student), Weihong Han, Xiaojiang Du, Mohsen Guizani, Zhihong Tian
    Simulation Modelling Practice and Theory (JCR-Q1)
    🔗 [Paper] 💻 [Code]

🛠️ Projects

  • DatalogBench: a benchmark for text-to-Datalog synthesis containing 136 tasks across four domains, each with a natural-language question, a relation schema, a reference program, and held-out input/output variants.

  • xAST: an open-source framework for systematically evaluating application security testing tools, providing standardized evaluation items and test cases to enable objective comparison and further improvement of the tools.

  • Bold-Falcon: an open-source cloud sandbox for automatically analyzing malware, which conducts a AI-enhanced detection module, providing comprehensive analysis results of malware input.

💬 Talks

  • 2025.12, Understanding Industry Perspectives of Static Application Security Testing (SAST) Evaluation. In Guangzhou University at Guangzhou, China.
  • 2025.06, Understanding Industry Perspectives of Static Application Security Testing (SAST) Evaluation. In the FSE’25 Security Session at Trondheim, Norway. 🔗 [slides]

🧑‍💻 Experiences

  • 2026.03-2027.03, Software Systems Engineering Group at University College London (UCL), London, United Kingdom.
  • 2022.06-2022.09, Mac Threat Response Team at Trend Micro, Nanjing, China.
  • 2018.05-2018.12, Information Security Technology Department at Cyberguard Technologies, Weihai, China.

🎖 Honors and Awards

  • 2025.12 Zhejiang University Qiushi Feiying Program
  • 2023.06 Outstanding Graduate
  • 2022.12 National Scholarship
  • 2021.12 Second Prize, CCF BCDI Digital Security Contest
  • 2020.10 Excellent Award, Coremail Email Security Contest
  • 2018.12 Outstanding Undergraduate