👋 About me

Hello, I’m Chenyan, a third-year PhD candidate in Computer Science at the National University of Singapore (NUS). [CV]

I am supervised by Prof. Zhiyong Huang and Prof. Jin Song Dong at the Programming Languages and Software Engineering Lab (PLSE Lab) under NUS, and by Prof. Yun Lin at the CoPhi Research Group, Shanghai Jiao Tong University (SJTU).

My research focuses on AI for software engineering, with an emphasis on intelligent systems for understanding, predicting, and recommending code edits. I study how developers edit code in project-specific contexts, and I build benchmarks and models that improve code editing systems. Some of my research has been integrated into ByteDance’s Trae IDE.

🔥 News

• 2026.02:   🎉🎉 Our paper EditFlow: Benchmarking and Optimizing Code Edit Recommendation Systems via Reconstruction of Developer Flows has been accepted by OOPSLA’26.
• 2025.09:   🎉🎉 Our paper Learning Project-wise Subsequent Code Edits via Interleaving Neural-based Induction and Tool-based Deduction has been accepted by ASE’25.
• 2024.12:   🎉🎉 I passed the qualifying examination.
• 2024.03:   🎉🎉 Our paper CoEdPilot: Recommending Code Edits with Learned Prior Edit Relevance, Project-wise Awareness, and Interactive Nature has been accepted by ISSTA’24.

(^# refers to equal contribution, ^* refers to corresponding author)

OOPSLA'26

• Liu C, Lin Y^*, Chang J, Liu J, Qi B, Jiang B, Huang Z, Dong JS. EditFlow: Benchmarking and Optimizing Code Edit Recommendation Systems via Reconstruction of Developer Flows[J]. Proc. ACM Program. Lang., 10, OOPSLA1, Article 141 (April 2026), 28 pages. [Homepage] [PDF] [GitHub]

ASE'25

• Liu C, Lin Y^*, Huang Y, Chang J, Qi B, Jiang B, Huang Z, Dong JS. Learning Project-wise Subsequent Code Edits via Interleaving Neural-based Induction and Tool-based Deduction[C]//2025 40th IEEE/ACM International Conference on Automated Software Engineering (ASE). IEEE, 2025: 1377-1389. [Homepage] [PDF] [GitHub]

ISSTA'24

• Liu C^#, Cai Y^#, Lin Y^*, Huang Y, Pei Y, Jiang B, Yang P, Dong JS, Mei H. CoEdPilot: Recommending Code Edits with Learned Prior Edit Relevance, Project-wise Awareness, and Interactive Nature[C]//Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis. 2024: 466-478 [Homepage] [PDF] [GitHub]

📖 Education

• 2023.08 - Present, Ph.D. in Computer Science, National University of Singapore (NUS).
• 2021.08 - 2023.03, M.S. in Computer Science, National University of Singapore (NUS).
• 2017.09 - 2021.06, B.S. in Computer Science, Huazhong University of Science and Technology (HUST).

👨‍💻 Internship

• 2025.02 - 2025.10, Large Language Model Algorithm Intern, Trae, ByteDance, Shenzhen, China.
  • Integrated LSP capabilities from the research prototype (ASE 2025) into Trae Cue’s production pipeline.
  • Proposed the mental-flow concept for modeling developer coding behavior; implemented the core algorithm for consolidating fragmented edit logs into coherent, low-noise edit sequences, helping Trae Cue Pro land the mental-flow idea in production, and completed the EditFlow paper (OOPSLA 2026) during the internship.

🎖 Honors and Awards

• 2025.12 Research Achievement Award, NUS.
• 2024.11 Second prize (ranked $3_{rd}$ out of 74 teams) in Prototype System Competition for Software Research Achievements, China Computer Federation (CCF).
• 2024.08 Research Achievement Award, NUS.
• 2021.06 Outstanding Undergraduate Student, HUST.

👨‍🏫 Teaching

• 2026 Spring, Teaching Assistant, CS4211 Formal Methods for Software Engineering, NUS.
• 2024 Fall, Teaching Assistant, CS5228 Knowledge Discovery and Data Mining, NUS.
• 2024 Spring, Teaching Assistant, CS4248 Natural Language Processing, NUS.

🏷 Patents

• 2025.08 Method, System, Apparatus, and Computer-Readable Storage Medium for Constructing Code Editing Datasets. China Invention Patent, Publication No. CN121070316A
• 2025.08 Method, System, Apparatus, and Computer-Readable Storage Medium for Determining Code Editing Recommendations. China Invention Patent, Publication No. CN121029146A.
• 2024.05 Method, System, Apparatus, and Computer-Readable Storage Medium for Code Editing. China Invention Patent, Publication No. CN119088381A.

💬 Invited Talks

• 2024.12, Workshop on LLM-based Code Agent, International Conference on Distributed Artificial Intelligence, Singapore.