About Me
I am a Ph.D. candidate in Artificial Intelligence at the Theory of Computation (ToC) Lab, Yonsei University, advised by Prof. Yo-Sub Han. My research focuses on enhancing the safety and transparency of large language models (LLMs) through detection and attribution techniques. In particular, I explore two complementary approaches: linguistic feature-based detection, which analyzes statistical differences in writing patterns between human- and LLM-generated content; and LLM watermarking, which embeds imperceptible signals into generated outputs to enable post hoc identification. I have developed detection and watermarking systems that generalize across multiple modalities (natural language and code) and languages (including English, Korean, Python, C, C++, and Java). Beyond detection, I am deeply interested in AI ethics, LLM guardrails, and the development of interpretable, trustworthy systems for real-world AI applications.
Research Interests
- Detection of LLM-generated content using linguistic features
- LLM watermarking for content attribution and provenance tracking
- AI safety, ethics, and transparency
- Korean natural language processing
- Code intelligence and programming language modeling
Recent News
- [Aug 2025] Presented KatFishNet as first author at ACL 2025 (Main) in Vienna, Austria.
- [July 2025] Paper accepted to Expert Systems with Applications.
- [March 2025] Paper accepted to Engineering Applications of Artificial Intelligence.