Jaeik Kim

I'm a Ph.D. student at AIDAS Lab, Seoul National University, advised by Jaeyoung Do.

I study how multimodal foundation models can acquire, combine, and adapt capabilities across perception, generation, and action.

Current directions: Unified multimodal models · Model adaptation and composition · Vision-language-action

My Thoughts

Email / Google Scholar / GitHub / LinkedIn

News

[2026/09] One paper has been accepted to Neurocomputing.
[2026/09] Two papers have been accepted to NeurIPS 2026.
[2026/06] One paper has been accepted to ECCV 2026.
[2026/04] Our paper, Medic-AD, has been selected as a CVPR 2026 Oral presentation!
[2026/03] Dynin-Omni is now live! Check out the Project Page.
[2025/09] Two papers have been accepted to NeurIPS 2025.
[2025/05] One paper has been accepted to ICML 2025.

Publications

	What Comes Next? Omni-StoryBench for Evaluating Story-Grounded Omnimodal Generation Sieun Hyeon, Yejoon Lee, Mintaek Lim, Woojin Kim, Jaeik Kim, and Jaeyoung Do^† arXiv preprint, 2026 paper
	Dual Path Semantic Segmentation with Locality Guidance Tokens Jaeik Kim^, Yoonjae Baek, Aecheon Jung, Hongyeob Kim, Soyun Choi, Youngeun Kim, Jaeyoung Do, and Sungeun Hong^† Neurocomputing* (Q1), 2026 paper
	When Predicting Nothing Beats SAM 3: Revisiting Evaluation in Video Object Segmentation Jihwan Hong, Woohyeon Park, Jaeik Kim, and Jaeyoung Do^† The 40th Annual Conference on Neural Information Processing Systems (NeurIPS), 2026 paper
	Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model Hoeun Lee, Jaeik Kim, Jusang Oh, Jinhyeok Kim, Geon Choi, Hyeonggeun Kim, Jaeyoung Do^† The 40th Conference on Neural Information Processing Systems (NeurIPS), 2026 project page / paper / code (coming soon) / model (coming soon)
	SUM: Unified Geometric Surgery on Spatio-Temporal Adaptation Vectors for Federated Class Incremental Learning Jaeik Kim^, Jaeyoung Do^† The 19th European Conference on Computer Vision (ECCV)*, 2026 project page / paper / code
	Dynin-Omni: Omnimodal Unified Large Diffusion Language Model Jaeik Kim^, Woojin Kim, Jihwan Hong, Yejoon Lee, Sieun Hyeon, Mintaek Lim, Yunseok Han, Dogeun Kim, Hoeun Lee, Hyunggeun Kim, Jaeyoung Do^† arXiv preprint*, 2026 project page / paper / code / model / demo
	Medic-AD: Towards Medical Vision-Language Model's Clinical Intelligence Woohyeon Park, Jaeik Kim, Sunghwan Steve Cho, Pa Hong, Woo Kyoung Jeong, Yoojin Nam, Nam-Joon Kim, Ginny Wong, Ka Chun Cheung, Jaeyoung Do^† The 43rd IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026 Oral 🏅 Award Candidate paper / code
	MMPB: It's Time for Multi-Modal Personalization Jaeik Kim^, Woojin Kim, Woohyeon Park, Jaeyoung Do^† The 39th Conference on Neural Information Processing Systems (NeurIPS)*, 2025 project page / paper / dataset
	Exploring and Leveraging Class Vectors for Classifier Editing Jaeik Kim^, Jaeyoung Do^† The 39th Conference on Neural Information Processing Systems (NeurIPS)*, 2025 project page / paper
	SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding Woohyeon Park, Woojin Kim, Jaeik Kim, Jaeyoung Do^† The 42nd International Conference on Machine Learning (ICML), 2025 project page / paper