
HUMANE Lab (Human-centered Language and Data Intelligence Lab) at Soongsil University conducts research on language-centered AI, including large language models, vision-language models, and agentic systems. Our research aims to address societal challenges, such as misinformation and bias, through human-centered AI technologies grounded in real-world data and domain knowledge.
숭실대학교 사람 중심 언어 및 데이터 지능(휴메인) 연구실은 거대 언어 모델, 시각 언어 모델, 에이전트 시스템 등 언어 중심 AI 기술을 연구합니다. 실제 데이터와 도메인 지식에 기반한 사람 중심 AI 기술을 연구하여 허위 정보와 편향 등 사회적 난제를 해결하는 것을 목표로 합니다.
We study language-centered AI, with particular attention to the human-centered challenges that arise in its real-world applications. Our areas of interest include:
Kunwoo Park / 박 건우 (kunwoo.park at ssu dot ac dot kr)
연구 협업, 산학 협력 등 다양한 형태의 협업을 환영합니다.
Two papers, led by Dahyun and Taegyun respectively, have been accepted to EMNLP 2026.
Feb 5th, 2026We, led by Jaeyoon and Yejun, won first place in the multimodal fact-checking challenge hosted by the FEVER workshop (co-located with EACL 2026).
Nov 14th, 2025Jaeyoon Jung presented his ongoing work on multimodal fact-checking at the Human-centered AI Workshop, co-located with CIKM 2025.
Oct 1st, 2025Dahyun Lee was awarded the Best Presentation Award at the HCLT 2025 (한글 및 한국어 정보처리 학술대회).
Aug 21th, 2025Our paper has been accepted to EMNLP 2025. We present an agentic in-context learning method that addresses long-context challenges in news stance detection, along with a novel dataset.
May 16th, 2025A paper has been accepted to ACL 2025 (Findings), uncovering potential data leakage in LLM-based query expansion and highlighting its limitations for real-world fact verification.
May 13th, 2025Building on last year’s success, we have once again secured second place in the fact-checking challenge at the FEVER workshop, co-located with ACL 2025.
Nov 11th, 2024Prof. Park won the Outstanding Reviewer Award at EMNLP 2024.
Sep 30th, 2024We placed runner-up at the AVeriTeC challenge, aiming to verify real-world claims using LLMs (collaboration with maum.ai and Adobe Research). Our system is invited as an oral presentation at the FEVER workshop co-located with EMNLP 2024.