I am an M.S. student at Korea University, co-advised by Prof. Donghyun Kim and Prof. Suhyun Kim at Kyung Hee University (formerly at KIST).

My research interest is to leverage image/video generation models to solve challenging and complex real-world tasks, especially those related to human perception. I am driven by the philosophy of “Generate to Understand”—echoing Feynman’s dictum, “What I cannot create, I do not understand”.

Beyond mere content creation, I aim to leverage generative models as fundamental instruments to solve complex real-world tasks. I believe that just as LLMs have evolved to reason and solve problems, vision generative models will ultimately serve as the key to perceiving, reasoning about, and understanding the physical world.

Email: climba (at) korea (dot) edu | Curriculum Vitae

📝 Publications

• Under Review Self-Improving Diffusion Classifiers with Minority Preference Optimization, Hyunsoo Kim*, Jungmyung Wi*, Soobin Um, Donghyun Kim†, Suhyun Kim†.

• Under Review Safety-Aware Image-to-Image Translation without Paired Data, Hyunsoo Kim, Wonjun Lee, Donghyun Kim, Suhyun Kim.

• ICML 2025 When Model Knowledge meets Diffusion Model: Diffusion-assisted Data-free Image Synthesis with Alignment of Domain and Class, Yujin Kim*, Hyunsoo Kim*, Hyunsoo J. Kim, Suhyun Kim (*: Equal contribution). Project page

• CVPR 2025 Difference Inversion: Interpolate and Isolate the Difference with Token Consistency for Image Analogy Generation, Hyunsoo Kim, Donghyun Kim†, Suhyun Kim†. Project page

💻 Work Experience

• 2025.09 - Present, Korea Institute of Science and Technology, Graduate Researcher, Seoul, Korea.
• 2024.03 - 2024.06, Korea Institute of Science and Technology, Undergraduate Research Intern, Seoul, Korea.

🤝 Academic Services

• Conference reviewer: AAAI, ICLR, CVPR

📖 Educations

• 2024.09 - Present, Korea University, M.S. in Computer Science (in progress).
• 2018.03 - 2024.08, Inha University, B.S. in Statistics.