He Ye | Assistant Professor at UCL

This is Ye. Welcome to my page. I am a Lecturer (aka. Assistant Professor) in the Department of Computer Science at University College London. Previously, I worked as a Postdoctoral Researcher at Carnegie Mellon University (2023 to 2025) with Prof. Claire Le Goues. I obtained my PhD from KTH Royal Institute of Technology in February 2023, where I was fortunate to be supervised by Prof. Martin Monperrus and Prof. Benoit Baudry.

My research aims to develop the next generation of AI code agents that can automate complex software engineering tasks. We study how large language models can understand large codebases, interact with development environments, and assist developers in solving real software problems. Our main research directions include:

• Benchmarking AI for software engineering: we build benchmarks and evaluation frameworks to measure how well AI models and agents perform real development tasks such as debugging and issue resolution.
• Codebase understanding and context retrieval: we develop methods that help AI systems locate the most relevant files, functions, and dependencies in large code repositories using structured representations such as knowledge graphs.
• Autonomous code agents and environment interaction: we build agents that can automate development workflows, including fixing issues, setting up environments, running terminal commands, and generating tests.
• Training and coordinating AI agents: we explore reinforcement learning and multi agent systems to train more reliable code agents that can reason, execute tasks, and collaborate to solve complex problems.

News

• [2025-11] Prometheus achieved TOP 5 and TOP 1 🏆 in the code agents using gpt-5 in the SWE-bench Verified leaderboard 🎉! Check out our Project Repository.
• [2025-11] Prometheus achieved TOP 1 🏆 by resolving 33.77% of issues in the AWS SWE-PolyBench leaderboard. Check out our Experiment Result.
• [2025-10] We maintain a curated list of code agent projects and research: Awesome Code Agents - explore the cutting-edge developments in autonomous code generation, bug fixing, and software engineering automation.

Team

PhD students

• Zhaoyang Chu (starting Fall 2025)
• Xiang Li (starting Fall 2025)
• Sean Xiangxin Zhao (starting Spring 2026)

Research Assistants

• Yue Pan

Visiting PhDs

• Zhiwei Fei (2025-2026) from Nanjing University
• Lingxiao Tang (2025-2026) from Zhejiang University
• Zhenyu Yang (2025-2026) from Shandong University

Publications

For a complete list, please visit my Google Scholar.

Code Agents

• Prometheus: Unified Knowledge Graphs for Issue Resolution in Multilingual Codebases.
  Zimin Chen, Yue Pan, Siyu Lu, Jiayi Xu, Claire Le Goues, Martin Monperrus, He Ye.
  Arxiv 2025.
• Understanding Code Agent Behaviour: An Empirical Study of Success and Failure Trajectories.
  Oorja Majgaonkar, Zhiwei Fei, Xiang Li, Federica Sarro, He Ye.
  Arxiv 2025.
• AdverIntent-Agent: Adversarial Reasoning for Repair Based on Inferred Program Intent.
  He Ye, Aidan ZH Yang, Chang Hu, Yanlin Wang, Tao Zhang, Claire Le Goues.
  Proceedings of the ACM on Software Engineering 2 (ISSTA), 1398–1420, 2025.
• MemoryBank: Enhancing Large Language Models with Long-Term Memory.
  Wanjun Zhong, Lianghong Guo, Qiqi Gao, He Ye, Yanlin Wang.
  Proceedings of the 38th AAAI Conference on Artificial Intelligence, 38(17), 19724–19731.

Model Post-Training and Reasoning

• CodeReasoner: Enhancing the Code Reasoning Ability with Reinforcement Learning.
  Lingxiao Tang, He Ye, Zhongxin Liu, Xiaoxue Ren, Lingfeng Bao.
  Arxiv 2025.
• Security vulnerability detection with multitask self-instructed fine-tuning of large language models.
  Aidan ZH Yang, Haoye Tian, He Ye, Ruben Martins, Claire Le Goues.
  Arxiv 2024.

Learning-based Automated Program Repair

• Iter: Iterative neural repair for multi-location patches.
  He Ye, Martin Monperrus.
  Proceedings of the 46th IEEE/ACM International Conference on Software Engineering, 2024.
• Neural program repair with execution-based backpropagation.
  He Ye, Matias Martinez, Martin Monperrus.
  Proceedings of the 44th International Conference on Software Engineering, 2022.
• SelfAPR: Self-supervised program repair with test execution diagnostics.
  He Ye, Matias Martinez, Xiapu Luo, Tao Zhang, Martin Monperrus.
  Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering, 2022.

Service

Organizer

• The 67th CREST Open Workshop (COW), UCL (2025)

  Workshop Chair

• Automated Program Repair @ ICSE (2025)
• Language Models and Program Languages @ SPLASH (2025)

  Program Committee Member

• ICSE Research Track (2025, 2026)
• ASE Research Track (2024, 2025)
• ESEC/FSE Research Track (2026)
• ISSTA Research Track (2026)
• MSR Research Track (2024)

  Organizing Committees

• Proceedings Co-chair, 15th International Conference on Internetware (2024)
• Publicity Co-chair, 31st Asia-Pacific Software Engineering Conference (2024)
• Doctoral Symposium Co-chair, 16th Asia-Pacific Symposium on Internetware (2025)
• Publicity Co-chair, 16th International Conference on Internetware (2025)

  Journal Reviewers

• IEEE TSE (2022–2025)
• ACM TOSEM (2023–2025)
• IEEE TDSC (2022–2024)
• Automated Software Engineering Journal (2024)
• Elsevier SCP (2023), Elsevier IST (2023)

  Workshop Reviewers

• LLM4Code @ ICSE (2025)
• AI-Augmented SDLC @ FSE (2025)
• AgenticSE @ ASE (2025)
• APSEC Demo Track (2025)

Contact

• Email: he[DOT]ye[AT]ucl[DOT]ac[DOT]uk