Our Research Focus: AI-Powered Software Engineering
We develop cutting-edge AI and software engineering techniques to solve real-world problems. Our core research involves building highly effective and adaptable RAG-sLLM and RAG-LMM. Key research topics include:
- LLM-Based Multi-Agent Bug Fixing: A modular framework where specialized sLLM agents collaborate to perform automated debugging.
- LLM-Based Bug Report Generation: Automating the creation of clear, reproducible, and context-rich issue descriptions.
- LLM-Based Test Case Generation: Generating and validating robust software test cases by analyzing requirements and code.
- LLM-Based Commit Message Generation: Automatically generating concise commit messages from code changes to improve repository history.
- LLM-Based Bug Fixing Optimization: Enhancing bug detection and patch generation accuracy under Zero-shot, Few-shot, and Fine-tuned settings.