My research interests have centered on computer vision and machine learning, with a special focus on action recognition—encompassing audio-visual integration, few-shot learning, and cost-efficient modeling approaches—as well as on transfer learning and personalization of deep models including LLM. I have had the privilege of being the first author on a number of significant research papers, which have been published in top-tier venues (ICCV, ICLR, AAAI, ICML, ACL etc.). Currently, my attention is directed towards cutting-edge research in Large Language Models (LLMs), especially RAG & on-device LLM.Homepage: https://sites.google.com/view/home-jtlee?pli=1