Embedded Files
CORE-AI LAb @ DGIST EECS
CORE-AI LAb @ DGIST EECS
CORE-AI Lab in the Department of Electrical Engineering and Computer Science (EECS) at Daegu Gyeongbuk Institute of Science and Technology (DGIST) develops collaborative, reliable, and efficient AI systems that operate across distributed devices, heterogeneous modalities, and real-world environments.
Our research focuses on building scalable AI systems that unify multimodal perception, representation learning, and real-world interaction. In particular, we study how AI models can learn compositional latent representations that generalize across heterogeneous modalities, structures, and environments. This enables adaptive intelligence for modern multimodal AI models such as vision-language models (VLMs), vision-language-action models (VLAs), and emerging physical AI systems.
Current research topics include:
Collaborative AI – developing collaborative learning frameworks such as federated and personalized learning
Efficient On-device AI – designing efficient inference and training frameworks for large models under device constraints
Reliable and Adaptive AI – improving reliability and controllability of AI systems through machine unlearning, task arithmetic, and hallucination mitigation
Multimodal AI – developing AI models that integrate heterogeneous modalities through modality fusion, representation alignment, and structured latent representations
Recent Updates
(Feb 2026) [New member] Doyoun, Myungbeom, Jooseok and Hyunwoo joined our lab, welcome!!
(Feb 2026) [New paper] Our paper "First Logit Boosting: Visual Grounding Method to Mitigate Object Hallucination in Large Vision-Language Models", has been accepted in CVPR 2026!! 🎉Congrats Jiwoo and Jongwoo!!
(Feb 2026) [New member] Dr. Malik Muhammad Saad joined our lab as a postdoc in InnoCORE program. Welcome!!
(Jan 2026) [New paper] Our paper "AROMMA: Unifying Olfactory Embeddings for Single Molecules and Mixtures" has been accpeted in IEEE ICASSP 2026!! 🎉Congrats Dayoung and Jongwon!!
(Dec 2025) [Research Grant] Our lab has been selected for funding by the Samsung Science & Technology Foundation (삼성미래기술육성사업), 2025-2029.
(Nov 2025) [New paper] Our paper "Stable Olfactory Receptor Activation across Odor Complexity" has been published in iScience!!
(Aug 2025) [New member] Two MS students (Jongwoo and Jinhwi) joined our lab. Welcome!!
(Jul 2025) [Research Grant] Our lab has been selected for funding by MSIT AI Star Fellowship (AI 스타펠로우십 사업), 2025-2030.
(Apr 2025) [Research Grant] Our lab has been selected for funding by MSIT InnoCORE program (InnoCORE 사업), 2025-2029.
(Feb 2025) [New member] Five MS students (Jiwoo, Jongwon, Jinwook, Dayoung, and Hangyeol) joined our lab, welcome!!
(Apr 2024) [New paper] Our paper, "LightVeriFL: A Lightweight and Verifiable Secure Aggregation for Federated Learning" has been published in IEEE JSAIT!!
(Apr 2024) [Research Grant] Our lab has been selected for funding by NRF Steam R&D Project (STEAM연구사업), 2024-2028.
(Apr 2024) [Research Grant] Our lab has been selected for funding by MSIT Regional Digital Innovation Cluster Support Project (지역 디지털 혁신거점 조성지원 사업), 2024-2026.
For Prospective Students
We always looking for highly-motivated undergraduate/graduate students. If you want to working with us, please apply to the Graduate program of the EECS Department and Interdisciplinary Studies of Atrificial Intelligence (AI) at DGIST and send me an email(jinhyun@dgist.ac.kr) with the title "Prospective Student" along with your CV, transcript, and research interests. Students who have majored in math with no engineering background are also welcome to apply.
Distributed AI Lab에서는 열정을 가진 학생을 모집하고 있습니다.
지원학과: 전기전자컴퓨터공학과 및 학제학과 인공지능전공
학위과정: 학부생인턴, 석사과정, 석박통합과정, 박사과정
지원시기: 상시(학부생인턴), 봄학기 1차,2차,3차 또는 가을학기(그 외)
연구실 진학에 관심 있으신 분은 언제나 면담 가능하오니, 제목에 "Prospective Student"를 기입하고 CV, 관심연구분야 및 성적표를 첨부하여 jinhyun@dgist.ac.kr 로 메일 주시기 바랍니다.
Page updated
Google Sites
Report abuse