Ph.D. Student / Machine Learning Researcher

Taeheon Kim

I am a 2nd-year Ph.D. student in Electrical and Computer Engineering at Machine Perception and Reasoning Lab, Seoul National University, advised by Prof. Jonghyun Choi.

My research focuses on model personalization through efficient learning, including continual learning, domain adaptation, machine unlearning, and knowledge transfer such as federated learning and model merging.

I received my B.S. in Computer Science from Yonsei University under the supervision of Prof. Jonghyun Choi.

CV Email Google Scholar LinkedIn GitHub

Latest

News

  1. DART [4] has been accepted to ECCV 2026.
  2. FedMosaic [3] received the Best Paper Award at the AAAI 2026 PerFM Workshop.
  3. FedMosaic [3] has been accepted to ICLR 2026.
  4. IDI [2] has been accepted to AAAI 2026.
  5. Our workshop proposal, CLVISION: Continual Learning in Computer Vision, has been accepted to ICCV 2025. I will serve as a Challenge Chair for ICCV 2025 @ CLVISION.
  6. RepBend [1] has been accepted to ACL 2025 Main.
  7. I started my Ph.D. at Seoul National University under the supervision of Prof. Jonghyun Choi.

Selected Work

Publications

Domain Arithmetic teaser

[4] Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shifts

Taewook Kang*, Taeheon Kim*, Donghyun Shin, Jonghyun Choi

ECCV 2026 to appear

Representation Bending teaser

[1] Representation Bending for Large Language Model Safety

Ashkan Yousefpour*, Taeheon Kim*, Ryan S. Kwon, Seungbeen Lee, Wonje Jeung, Seungju Han, Alvin Wan, Harrison Ngan, Youngjae Yu, Jonghyun Choi

ACL 2025 Main

PDF BibTeX Code

* indicates equal contribution.

Recognition

Awards