avatar

Jiwoo Chung

Ph.D. student
Sungkyunkwan University
jiwoo.jg@gmail.com


About Me

I am a Ph.D. student at the Visual Computing Lab, Sungkyunkwan University (SKKU), advised by Prof. Jae-Pil Heo. I received my M.S. and B.S. degrees from SKKU. My research interests lie in Generative Models, with a recent focus on world models and diffusion model acceleration. Previously, I interned at NAVER Cloud (2025.05 - 2025.09), where I worked on accelerating video diffusion models using caching. I am currently seeking research internship opportunities.

News

Publications

  1. SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Models
    Jiwoo Chung, Sangeek Hyun, MinKyu Lee, Byeongju Han, Geonho Cha, Dongyoon Wee, Youngjun Hong*, and Jae‑Pil Heo* (*: Co‑corresponding authors.)
    CVPR 2026 Oral & Best Paper Finalist (Top 0.1%, 15/16092)
    Arxiv Code Project page
  2. Analyzing the Training Dynamics of Image Restoration Transformers: A Revisit to Layer Normalization
    MinKyu Lee, Sangeek Hyun, Woojin Jun, Hyunjun Kim, Jiwoo Chung, and Jae‑Pil Heo
    ICLR 2026
    OpenReview Arxiv Code
  3. Diffusion Feature Field for Text-based 3D Editing with Gaussian Splatting
    Eunseo Koh, Sangeek Hyun, MinKyu Lee, Jiwoo Chung, Kangmin Seo, and Jae‑Pil Heo
    NeurIPS 2025
    OpenReview
  4. Fine-Tuning Visual Autoregressive Models for Subject-Driven Generation
    Jiwoo Chung, Sangeek Hyun, Hyunjun Kim, Eunseo Koh, MinKyu Lee, and Jae‑Pil Heo
    ICCV 2025
    Qualcomm Innovation Fellowship Korea 2025 Finalist
    Paper Arxiv Code Project page
  5. Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer
    Jiwoo Chung*, Sangeek Hyun*, and Jae‑Pil Heo (*: equal contribution)
    CVPR 2024 Highlight (Top 2.8%, 324/11532)
    Paper Arxiv Code Project page
  6. Diversity-aware Channel Pruning for StyleGAN Compression
    Jiwoo Chung, Sangeek Hyun, Sang‑Heon Shim, and Jae‑Pil Heo
    CVPR 2024
    Paper Arxiv Code Project page
  7. Towards Squeezing-Averse Virtual Try-On via Sequential Deformation
    Sang‑Heon Shim, Jiwoo Chung, and Jae‑Pil Heo
    AAAI 2024
    Arxiv Code
  8. Frequency-based Motion Representation for Video Generative Adversarial Networks
    Sangeek Hyun, Jaihyun Lew, Jiwoo Chung, Euiyeon Kim, and Jae‑Pil Heo
    TIP 2023
    PDF Code

Reviewer


Powered by Jekyll and Minimal Light theme.