Our lab is currently seeking highly motivated undergraduate research students (senior-year or above) who are planning to pursue graduate studies ๐
We particularly welcome students with a strong interest in generative language models, natural language processing, or machine learning. This position is well suited for students who are eager to gain hands-on research experience and build a strong foundation for graduate-level study.
If you are interested, please send a brief self-introduction and your academic background to sh0416{at}knu.ac.kr.
For more information, please refer to this post.
For GKS Scholarship Applicants, please refer to this post.
Our current research focuses on augmenting generative language models as effective coding agents ๐ค.
In particular, we study methods that enable language models to reason over code, interact with external tools, and leverage structured context to solve complex programming tasks. Our goal is to improve the reliability, controllability, and practical usefulness of code-generating agents in real-world development settings.
This includes, but is not limited to, research on prompt and context design, tool-augmented generation, code understanding and editing, and evaluation methodologies for coding agents.
Our primary research project is supported by a government-funded initiative on developing code-expert language models and coding agents for AI-assisted software development. This project aims to move beyond line-level or function-level code generation and instead focuses on language models that can understand and operate over entire code repositories.
Specifically, we study problems such as repository-level code completion, context construction across multiple files, long-range dependency modeling in code, and agentic workflows for software development tasks. Our research explores how language models can navigate large codebases, reason about project structure, and iteratively interact with tools to support realistic development scenarios.