Taehoon Yoon

I am a Ph.D. student at the University of Michigan, advised by Prof. Liyue Shen. Previously, I completed my M.S. in Artificial Intelligence at KAIST under the supervision of Prof. Minhyuk Sung. I was also fortunate to be mentored by Prof. Jong Chul Ye at KAIST during my research internship. I received my B.S. in Physics with a double major in Computer Science and Engineering from Sogang University, where I graduated as valedictorian of the College of Natural Sciences.

My research interests broadly lie in generative modeling and probabilistic inference. I am particularly interested in modern generative modeling frameworks, including diffusion models, flow-matching models, and other emerging generative modeling paradigms. From the perspective of probabilistic inference, I am especially interested in sampling methods and their connections to modern generative models. A central theme of my research is to improve the controllability, alignment, and reliability of generative models at inference time, with a focus on controllable generation, reward alignment, inverse problems, and particle-based sampling methods.

I am always open to collaborations, so please feel free to contact me!