ICLR
2026
OmniText: A Training-Free Generalist for Controllable Text-Image Manipulation
* Equal contribution · † Co-corresponding authors · International Conference on Learning Representations, 2026
About
I am a Ph.D. student at the Video and Image Computing Lab (VIC Lab), KAIST, advised by Prof. Munchurl Kim. Before joining KAIST, I worked as an R&D engineer at the AI Center of China Medical University Hospital in Taiwan, where I built machine-learning systems for clinical decision support, including an AI system for sepsis detection and mortality prediction deployed in hospital practice. I received my M.S. in Electrical Engineering from City University of Hong Kong and my B.S. from Southeast University (Nanjing, China).
My research interests include image editing and controllable generative models, as well as skeleton representation learning and motion generation.
News
Publications
International Conferences
Preprints
arXiv
2026
MotionGrounder: Grounded Multi-Object Motion Transfer via Diffusion Transformer
arXiv preprint, 2026
arXiv
2026
Under Review
Under Review
2026
MotionTok: What Makes a Good Motion Tokenizer for Motion Generation?
* Equal contribution · Under review, 2026
Under Review
2026
One for All: A Generalist Foundation Model for Cross-Sensor Skeleton Representation Learning
Under review, 2026
Domestic Conferences
KIBME
2026
Skeleton Aware Augmentation for Skeleton Representation Learning
Korea Institute of Broadcast and Media Engineers (KIBME) Workshops, Jun. 2026
KIBME
2025
Attention Modulation for Improved Textual Alignment in Motion Transfer
Korea Institute of Broadcast and Media Engineers (KIBME) Workshops, Dec. 2025
Experience
- Assistant R&D Engineer, AI Center, China Medical University Hospital Clinical AI systems: sepsis detection & mortality prediction (“i.A.M.S”), bone-age estimation, early-warning and infection-risk models · Taichung, Taiwan
Education
- Integrated M.S.–Ph.D., School of Electrical Engineering, KAIST Video and Image Computing Lab · Advisor: Prof. Munchurl Kim
- M.S., Department of Electrical Engineering, City University of Hong Kong Thesis: Grading for Prostate Cancer with Deep Learning · Multimedia Information Technology
- B.S., School of Information Science and Engineering, Southeast University Thesis: Research on Neural Network Compression Method · Information Engineering · Nanjing, China
Honors & Awards
- Minor Merit & Commendation, China Medical University Hospital, 2023
- YOUNG Awards — Smart Application, Champion, 2022
- Symbol of National Quality (SNQ) — Bronze Medal, 2022
- National Healthcare Quality Award (NHQA) — Honorable Mention, 2022
Patents
- AI Assisted Medical Diagnosis Method for Sepsis and System Thereof
Taiwan patent application · China Medical University Hospital
Languages
- Chinese (native) · English (IELTS 7.5) · Korean (TOPIK Level 5)