Dongwon Kim

Dongwon Kim

world model / representation learning / multi-modal learning

I am a postdoctoral researcher at KAIST, working with Prof. Jeany Son. I previously completed my BS and PhD at POSTECH CVLab where I worked with Prof. Suha Kwak. My research centers on whether machines can learn representations with the right abstraction and hierarchy, spanning work in compositional representation (SelfMod), multi-modalities (DivE, SaG, MaskGen), and world model (ongoing, CompACT).

News 01 / 13

24 Sep 2026 A paper on generation-friendly image tokenization is accepted at NeurIPS 2026 (arXiv).

all news →

Publications

  1. arXiv 2026 ACID: Action Consistency via Inverse Dynamics for Planning with World Models WM
    ACID: Action Consistency via Inverse Dynamics for Planning with World Models preview
    Gawon Seo, Dongwon Kim, and Suha Kwak
    arXiv 2026
  2. CVPR 2026 Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model WM·GEN
    Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model preview
    Dongwon Kim, Gawon Seo, Jinsung Lee, Minsu Cho, and Suha Kwak
    CVPR 2026
    (An early version appeared in LSRW workshop in CoRL 2025)
  3. NeurIPS 2026 Structured State-Space Regularization for Generation-Friendly Image Tokenization GEN
    Structured State-Space Regularization for Generation-Friendly Image Tokenization preview
    Jinsung Lee, Jaemin Oh, Namhun Kim, Dongwon Kim, Byung-Jun Yoon, and Suha Kwak
    NeurIPS 2026
  4. ICCV 2025 Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens GEN·V+L
    Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens preview
    Dongwon Kim*, Ju He*, Qihang Yu*, Chenglin Yang, Xiaohui Shen, Suha Kwak, and Liang-Chieh Chen
    ICCV 2025
  5. arXiv 2024 1.58-bit FLUX GEN
    1.58-bit FLUX preview
    Chenglin Yang, Celong Liu, Xueqing Deng, Dongwon Kim, Xing Mei, Xiaohui Shen, and Liang-Chieh Chen
    arXiv 2024
  6. NeurIPS 2024 Bootstrapping Top-down Information for Self-modulating Slot Attention REP
    Bootstrapping Top-down Information for Self-modulating Slot Attention preview
    Dongwon Kim, Seoyeon Kim, and Suha Kwak
    NeurIPS 2024
  7. ECCV 2024 PLOT: Text-based Person Search with Part Slot Attention for Corresponding Part Discovery V+L
    PLOT: Text-based Person Search with Part Slot Attention for Corresponding Part Discovery preview
    Jicheol Park, Dongwon Kim, Boseung Jeong, and Suha Kwak
    ECCV 2024
  8. NAACL 2024 Extending CLIP’s Image-Text Alignment to Referring Image Segmentation V+L
    Extending CLIP’s Image-Text Alignment to Referring Image Segmentation preview
    Seoyeon Kim, Minguk Kang, Dongwon Kim, Jaesik Park, and Suha Kwak
    NAACL 2024
  9. ICCV 2023 Shatter and Gather: Learning Referring Image Segmentation with Text Supervision V+L
    Shatter and Gather: Learning Referring Image Segmentation with Text Supervision preview
    Dongwon Kim*, Namyup Kim*, Cuiling Lan, and Suha Kwak
    ICCV 2023
  10. CVPR 2023 Improving Cross-Modal Retrieval With Set of Diverse Embeddings V+L
    Improving Cross-Modal Retrieval With Set of Diverse Embeddings preview
    Dongwon Kim, Namyup Kim, and Suha Kwak
    CVPR 2023Highlight, 235/9155 = 2.5%
  11. CVPR 2022 ReSTR: Convolution-Free Referring Image Segmentation Using Transformers V+L
    ReSTR: Convolution-Free Referring Image Segmentation Using Transformers preview
    Namyup Kim, Dongwon Kim, Cuiling Lan, Wenjun Zeng, and Suha Kwak
    CVPR 2022
  12. CVPR 2022 Self-Taught Metric Learning Without Labels REP
    Self-Taught Metric Learning Without Labels preview
    Sungyeon Kim, Dongwon Kim, Minsu Cho, and Suha Kwak
    CVPR 2022
  13. CVPR 2021 Embedding Transfer With Label Relaxation for Improved Metric Learning REP
    Embedding Transfer With Label Relaxation for Improved Metric Learning preview
    Sungyeon Kim, Dongwon Kim, Minsu Cho, and Suha Kwak
    CVPR 2021
  14. CVPR 2020 Proxy Anchor Loss for Deep Metric Learning REP
    Proxy Anchor Loss for Deep Metric Learning preview
    Sungyeon Kim, Dongwon Kim, Minsu Cho, and Suha Kwak
    CVPR 2020

Experience

2025 – Now Postdoctoral researcher · KAIST, Daejeon, KR
2024 Research Intern · Fundamental Research Team, ByteDance SEED, San Jose, US
Developed efficient text-to-image generative model using 1D tokens (MaskGen)

Education

2019 – 2025 Integrated M.S & Ph.D in Computer Science & Engineering
POSTECH, Pohang, South Korea
Advisor: Prof. Suha Kwak
2015 – 2019 B.S. in Computer Science & Engineering
POSTECH, Pohang, South Korea

Honors and Awards

POSTECH CSE Best Research Award, POSTECH, 2025

POSTECHIAN Fellowship, POSTECH, 2023

  • $5,000 grant

BK21 Best Paper Award, POSTECH GSAI, 2023

  • Self-Taught Metric Learning without Labels (CVPR 2022)

Qualcomm Innovation Fellowship Winner, Qualcomm Korea Corp., 2022

  • Self-Taught Metric Learning without Labels (CVPR 2022)
  • ReSTR: Convolution-free Referring Image Segmentation Using Transformers (CVPR 2022)

NAVER x POSTECH AI DAY The 2nd and 3rd Prize, 2022

  • ReSTR: Convolution-free Referring Image Segmentation Using Transformers (CVPR 2022)

Qualcomm Innovation Fellowship Winner, Qualcomm Korea Corp., 2021

  • Embedding Transfer with Label Relaxation for Improved Metric Learning (CVPR 2021)

IPIU Best Paper Award, 2021

  • Embedding Transfer with Label Relaxation for Improved Metric Learning (CVPR 2021)

Professional Services

Reviewer
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
  • IEEE/CVF International Conference on Computer Vision (ICCV)
  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • European Conference on Computer Vision (ECCV)
  • Winter Conference on Applications of Computer Vision (WACV)
  • Asian Conference on Computer Vision (ACCV)
  • Conference on Neural Information Processing Systems (NeurIPS)