Minhyuk Seo

Computer Science Ph.D. Student · UC Berkeley EECS

I'm a Computer Science Ph.D. student in the Department of Electrical Engineering and Computer Sciences (EECS) at UC Berkeley, advised by Prof. Joseph E. Gonzalez .

Before joining UC Berkeley, I was a Research Associate at KU Leuven, working with Prof. Tinne Tuytelaars. I received my M.S. and B.S. degrees in Computer Science from Yonsei University, advised by Prof. Jonghyun Choi.

My research focuses on human-like AI, inspired by how humans learn and evolve. In particular, I focus on (i) continual learning and (ii) collaborative learning, with the goal of enabling human-like learning that continually acquires knowledge over time and evolves through interaction with others.

If you are interested in my research, feel free to reach out by email.

Continual Learning Collaborative Learning Multimodal AI

CV  /  Portfolio  /  Email  /  Linkedin  /  Google Scholar  /  DBLP  /  Github

profile photo
News
2026.09 Our ProCoRe paper was accepted to NeurIPS 2026.
2026.08 I started my Ph.D. in EECS at UC Berkeley, advised by Prof. Joseph E. Gonzalez .
2026.06 Our MoRAS paper was accepted to ECCV 2026.
2026.05 I recieved ICML 2026 Gold Reviewer Award.
2026.04 Our OASIS paper was accepted to ACL 2026 (main).
2026.01 Our Co-LoRA paper recieved Best Paper Award at AAAI 2026 workshop of PerFM .
2026.01 Our Co-LoRA paper was accepted to ICLR 2026.
2025.12 I am invited to give a talk at NYU Agentic Learning AI Lab - Human-like AI
2025.09 Our GenOL paper was accepted to TMLR 2025.
2025.07 I am invited to give a talk at Princeton Visual AI Lab - Toward Efficient Training of MLLMs
2025.05 Our workshop proposal (CLVISION: Continual Learning in Computer Vision) was accepted to ICCV 2025. I will serve as an Challenge Chair in ICCV 2025 @ CLVISION .
2025.03 I started as a Research Associate at KU Leuven under the supervision of Prof. Tinne Tuytelaars
2025.01 Our aL-SAR paper was accepted to ICLR 2025 (Spotlight Presentation)
2024.09 I am invited to give a talk at Naver Labs - Recent research on continual learning
2024.06 Two papers accepted to CoLLAs 2024 Workshop Track
2024.06 I am invited to give a talk at Apple research - Practical set‑ups and methods for continual learning
2024.02 Our EARL paper was accepted to CVPR 2024
2024.01 Our CL-ALFRED paper was accepted to ICLR 2024
Work Experience
KU Leuven
Research Associate at ESAT-PSI Lab (Mar. 2025 ~ Jul. 2026)
Advisor: Prof. Tinne Tuytelaars
Seoul National University
Researcher at SNUMPR Lab (Sep. 2024 ~ Feb. 2025)
Advisor: Prof. Jonghyun Choi
LG AI Research
Research Intern at AML Team (Feb. 2023 ~ Aug. 2023)
Mentor: Dr. Hankook Lee

Publications
Visual Grounding First, Multimodal In-context Learning Follows
Minhyuk Seo*, Minjae lee*, Chaeeun Lee, Wei Lin, Muhammad Jehanzeb Mirza, Tinne Tuytelaars† , Jonghyun Choi†
NeurIPS 2026
[ pdf ]
Attention Misses Visual Risk: Risk-Adaptive Steering for Multimodal Safety Alignment
Jonghyun Park, Minhyuk Seo , Chaewon Yeo, Jonghyun Choi
ECCV 2026
[ pdf ]
OASIS: Online Sample Selection for Continual Instruction Tuning
Minjae Lee* , Minhyuk Seo *, Tingyu Qu , Tinne Tuytelaars , Jonghyun Choi
ACL 2026 (main)
[ pdf ]
Co-LoRA: Collaborative Model Personalization on Heterogeneous Multi-Modal Clients
Minhyuk Seo *, Taeheon Kim*, Hankook Lee , Jonghyun Choi , Tinne Tuytelaars
ICLR 2026 / Best Paper Award @ AAAI 2026 Workshop of PerFM
[ pdf ]
GenOL: Generating Diverse Examples for Name Only Online Learning
Minhyuk Seo , Seongwon Cho, Minjae Lee, Diganta Misra, Hyeonbeom Choi, Seon Joo Kim , Jonghyun Choi ,
TMLR 2025
[ pdf ]
Budgeted Online Continual Learning by Adaptive Layer Freezing and Frequency-based Sampling
Minhyuk Seo *, Hyunseo Koh*, Jonghyun Choi ,
ICLR 2025 (Spotlight Presentation)
[ pdf ] [ bibtex ]
Learning Equi-angular Representations for Online Continual Learning
Minhyuk Seo Hyunseo Koh, Wonje Jeung, Min Jae Lee, San Kim, Hankook Lee , Sungjun Cho, Sungik Choi, Hyunwoo Kim, Jonghyun Choi ,
CVPR 2024
[ pdf ] [ bibtex ] [ code ]
Online Continual Learning for Interactive Instruction Following Agents
Byeonghwi Kim *, Minhyuk Seo *, Jonghyun Choi
ICLR 2024
[ pdf ] [ bibtex ] [ code ]
Online Boundary-Free Continual Learning by Scheduled Data Prior
Hyunseo Koh, Minhyuk Seo , Jihwan Bang , Hwanjun Song , Deokki Hong, Seulki Park, Jung-Woo Ha , Jonghyun Choi
ICLR 2023
[ pdf ] [ bibtex ] [ code ]
Awards
  • Travel Grant - The IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2025
  • Travel Grant - International Conference on Learning Representations (ICLR) 2025
  • 2nd Place Award - Class-Incremental with Repetition Scenario Challenge - CLVISION Workshop @ CVPR 2024
  • Bronze Prize - Outstanding Paper Awards - IPIU 2024
  • 1st Place Award - Continual Test-time Adaptation for Object Detection Challenge - VCL Workshop @ ICCV 2023
  • 1st Place Award (Minister of Science, ICT and Future Planning Award) - HANIUM ICT competition 2019
Academic Services
  • Reviewer
    • CVPR (2024-2026), ICML (2024-2026), ICLR (2024-2027), AISTATS (2025-2026), ICRA (2025), TPAMI (2024-2026), NeurIPS (2024-2026), IJCV (2025-2026), ICCV (2025), WACV (2026), ECCV (2026), AAAI (2026-2027), ACL ARR (May 2025), CoRL @ LLHomeRobots (2024)
  • Program committee member


This template is from Jon Barron .