Avatar

Joonsung Jeon

Ph.D. Candidate

School of Computing, KAIST


Google Scholar | GitHub


I am a first-year Ph.D. candidate in the Scalable Graphics, Vision, and Robotics (SGVR) Lab at KAIST, advised by Prof. Sung-Eui Yoon. I am broadly interested in the safety, privacy, and trustworthiness of AI systems, with a focus on generative and embodied AI.

Research Interests:

  • Adversarial attacks & robustness in ML
  • Membership inference attacks on generative models
  • Privacy and copyright protection in image generation
  • Safety and reliability in Vision-Language-Action (VLA) systems

Contact

  • mikeraph [at] kaist.ac.kr

  • Bldg E3-1, Rm 3443, 291 Daehak-ro, Yuseong-gu, Daejeon, Korea, 34141

Education

  • Ph.D. in School of Computing, KAIST

    - Advisor: Sung-Eui Yoon

    Mar. 2026 - Current

  • M.S. in Kim Jaechul Graduate School of AI, KAIST

    - Advisor: Sung-Eui Yoon

    Mar. 2024 - Feb. 2026

  • B.S. in Electrical and Electronic Engineering, Yonsei University

    Mar. 2018 - Feb. 2024

News

  • [Jan 2026] ✨ One paper is accepted to ICLR 2026!
  • [Jan 2025] One paper is accepted to ICLR 2025.
  • [Oct 2024] One paper is accepted to APL Photonics 2024.

Publications

  • No Caption, No Problem: Caption-Free Membership Inference via Model-Fitted Embeddings

    Joonsung Jeon, Woo Jae Kim, Suhyeon Ha, Sooel Son*, and Sung-eui Yoon*

    The International Conference on Learning Representations, 2026

    [Paper] [Code] [Project]

  • AdvPaint: Protecting Images from Inpainting Manipulation via Adversarial Attention Disruption

    Joonsung Jeon, Woo Jae Kim, Suhyeon Ha, Sooel Son*, and Sung-eui Yoon*

    The International Conference on Learning Representations, 2025

    [Paper] [Code] [Project]

  • Restoring H&E stain in faded slides via phase-to-color virtual staining in near-infrared

    H. Chae, J. Kim, J. Jeon, K. Lee, K. Lee, J. Choi, S. Kang, S. Choi, G. Bang, J. Lee, E. Park, N. Cho, and S. Lee

    APL Photonics, 2024

    [Paper]

  • Edge driven unsupervised phase-to-H&E Virtual Staining

    Jongho Kim, Hyesuk Chae, Joonsung Jeon, Kyung Chul Lee, and Seung Ah Lee

    Conference on Lasers and Electro-Optics/Pacific Rim, 2024

    [Paper]

  • Virtual Re-staining of Faded H&E-Stained Slides Using NIR Quantitative Phase Imaging

    Hyesuk Chae, Joonsung Jeon, Kyung Chul Lee, Ji Ung Choi, Kyungwon Lee, and Seung Ah Lee

    Imaging Systems and Applications, 2023

    [Paper]