Yongchan Chun

I recently earned my M.S. in Computer Science and Engineering at Korea University, where I was a member of the NLP&AI Lab advised by Heuiseok Lim. My research focuses on improving the reliability of AI models, with an emphasis on:

Improving model interpretability to better understand model behavior and guide modifications
Calibrating model confidence so that certainty aligns more closely with actual correctness

I welcome opportunities for discussion and collaboration. Please feel free to contact me via email or LinkedIn!

news

Apr 13, 2026	Two papers has been accepted to ACL 2026!
Feb 22, 2026	One paper on uncertainty estimation has been accepted to CVPR 2026. See you in Denver 🇺🇸!
Feb 02, 2026	I will be joining LG AI Research as a Research Intern, where I will work on developing foundation LLMs.
Sep 15, 2025	Two papers on LLM interpretability and modification have been accepted to EMNLP 2025. See you in Suzhou 🇨🇳!
Jun 25, 2025	Our paper on enhancing information extraction has been accepted to ACL Findings 2025. See you in Vienna 🇦🇹!

selected publications

CVPR Highlight

Evidential Transformation Network: Turning Pretrained Models into Evidential Models for Post-hoc Uncertainty Estimation

Yongchan Chun, Chanhee Park, Jeongho Yoon, and 2 more authors

2026
ACL Findings

Enhancing Automatic Term Extraction with Large Language Models via Syntactic Retrieval

Yongchan Chun, Minhyuk Kim, Dongjun Kim, and 2 more authors

In Findings of the Association for Computational Linguistics: ACL 2025 , Jul 2025

Abs

Automatic Term Extraction (ATE) identifies domain-specific expressions that are crucial for downstream tasks such as machine translation and information retrieval. Although large language models (LLMs) have significantly advanced various NLP tasks, their potential for ATE has scarcely been examined. We propose a retrieval-based prompting strategy that, in the few-shot setting, selects demonstrations according to syntactic rather than semantic similarity. This syntactic retrieval method is domain-agnostic and provides more reliable guidance for capturing term boundaries. We evaluate the approach in both in-domain and cross-domain settings, analyzing how lexical overlap between the query sentence and its retrieved examples affects performance. Experiments on three specialized ATE benchmarks show that syntactic retrieval improves F1-score. These findings highlight the importance of syntactic cues when adapting LLMs to terminology-extraction tasks.