Embedded Files
"AI For Good"
OUR Mission: Understanding, Updating, Repairing Knowledge in AI Models
OUR Mission: Understanding, Updating, Repairing Knowledge in AI Models
Our group conducts research on machine learning and data-centric AI problems & systems grounded in big data, spanning areas such as machine learning, data mining, databases, and computational theory. Rather than focusing solely on building ever-larger models, we study how learned knowledge inside modern AI models can be understood, reused, adapted, and trusted.
Our core research theme is to make large-scale AI models (particularly deep neural networks) more accessible, scalable, and reliable. In particular, we investigate the following questions:
(Accessibility) How can the knowledge learned by trained neural networks be extracted, reused, and explained?
(Scalability) How can AI models be continually and incrementally updated without forgetting previously learned knowledge?
(Reliability) How can incorrectly learned or biased knowledge be detected, repaired, or reorganized within neural networks?
From an academic perspective, we pursue research of the highest quality in both theory and practice. Our work aims to produce rigorous and impactful contributions, and we actively publish in top-tier venues in machine learning and big data, including ICML, NeurIPS, ICLR, AAAI, KDD, ICDM, SIGMOD, VLDB, ICDE, etc. For a comprehensive list of top-tier conferences and journals, please refer to the following Google Scholar pages: Artificial Intelligence, Data Mining & Analysis, and Databases & Information Systems.
LATEST NEWS
LATEST NEWS
[Jan 2026] A paper has been accepted in Journal of Artificial Intelligence Research (JAIR), a prestigious journal in the field of AI.
[Jan 2026] A paper has been accepted in Journal of Artificial Intelligence Research (JAIR), a prestigious journal in the field of AI.
Our paper, "PGB: One-Shot Pruning for BERT via Weight Grouping and Permutation ," (1st author: Hyemin Lim, MS alumna) has been accepted for publication in Journal of Artificial Intelligence Research, which is a prestigious SCIE journal (IF 4.6) in artificial intelligence. Well deserved, Hyemin & Jaeyeon!
[Jan 2026] A paper has been accepted in WWW 2026, a top-tier conference (BK 4) in the field of (Web) Data Mining & AI.
[Jan 2026] A paper has been accepted in WWW 2026, a top-tier conference (BK 4) in the field of (Web) Data Mining & AI.
Our paper, "STARK: Structure-Aware and Adaptive Representation Learning for Continual Knowledge Graph Embedding ," (1st author: Kyung-Hwan Lee, MS Student) has been accepted for the WWW 2026 (ACM The Web) conference, which is a top-tier conference in the fields of web-based data mining and AI (BK 4). Congratulations, Kyung-Hwan!
[Nov 2025] A paper has been accepted in ACM SAC 2026, a well-recognized international conference.
[Nov 2025] A paper has been accepted in ACM SAC 2026, a well-recognized international conference.
Our paper, "Balanced Online Class-Incremental Learning via Dual Classifiers ," (1st author: Shunjie Wen, PhD Student) has been accepted for the ACM SAC 2026 conference, which is a reputable international venue in applied computing research (BK 1). Congratulations, Shunjie!
[Jul 2025] A paper has been accepted in BMVC 2025, a prestigous Computer Vision conference annually held in the UK.
[Jul 2025] A paper has been accepted in BMVC 2025, a prestigous Computer Vision conference annually held in the UK.
Our paper, "FaceGCD: Generalized Face Discovery via Dynamic Prefix Generation," (1st author: Yunseok Oh, MS alumnus) has been accepted for publication in BMVC, which is a presitigious Computer Vision conference held annually in the UK. Congratulations, Yunseok!
[Jul 2025] A paper has been accepted in ECAI 2025, the best AI conference in Europe.
[Jul 2025] A paper has been accepted in ECAI 2025, the best AI conference in Europe.
Our paper, "Lossless Token Merging Even Without Fine-Tuning in Vision Transformers," (1st author: Jaeyeon Lee, PhD student) has been accepted for publication in ECAI, which is known as one of top 3 general AI conferences in the world and the best in Europe. Congratulations, Jaeyeon!
Vacancies
Vacancies
We are always looking for motivated and curious students (BS, MS, and PhD) who are interested in machine learning, data-centric AI, and large-scale models. Please email Prof. Choi your CV, transcript, and (if available) a brief statement of your research interests.
빅데이터 연구실에서는 학부연구생, 석사과정, 박사과정 학생을 모집하고 있습니다. 대학원 진학 및 연구실에 대해서 관심이 있는 학생들은 dchoi@inha.ac.kr로 CV와 간단한 연구(학업)계획서를 첨부해서 이메일을 보내기 바랍니다.
Google Sites
Report abuse