SYSU · SYSUSELab

Research

Research

My research lies broadly in LLM4SE, AI4SE, and intelligent software engineering. I am interested in how AI can support the entire software lifecycle, from understanding and generating code to testing, documenting, and maintaining large software systems.

Research Themes

  1. 01

    LLM4SE & AI4SE

    Applying large language models and AI methods across the software lifecycle, from requirements and design to coding, testing, and maintenance.

  2. 02

    Repository-Level Code Intelligence

    Understanding entire repositories through dependency-aware code graphs, structured context, and knowledge representations of code and documentation.

  3. 03

    Intelligent Software Maintenance

    Automated documentation, change impact analysis, program repair, and incremental updates for evolving software systems.

  4. 04

    Agentic & Knowledge-Graph-Powered Tools

    Combining LLM agents, tool use, knowledge graphs, and program analysis to build practical software engineering tools.

Across these directions, I am particularly interested in whether AI-powered software tools are reliable, useful, and practical in real development workflows.