I am Chanhyuk (David) Lee, M.S student at KAIST School of Computing. I’m currently working in VL Lab, which is led by Prof. Seunghoon Hong. Before joining the lab, I received my Bachelor’s degree from KAIST in a double major of Computer Science and Chemistry.
My research focuses on building fast and efficient generative models. Broadly, my work spans model merging and few-shot learning for generative models. Currently, I am working on non-autoregressive text generation through diffusion models, enabling parallel decoding of language. In particular, I am interested in continuous representations of text for diffusion language models, which make various techniques from the image domain, such as flow map distillation, directly transferable. I also worked on model merging ans few-shot learning, including studies on the parameter space geometries of neural networks.
News
2026.05
One paper is accepted to ICML 2026.
2026.01
One paper is accepted to ICLR 2026.
2025.09
One paper is accepted to NeurIPS 2025.
Publications
PatchKV: Weight-Space Compensation of KV Cache
Chanryeol Lee, Chanhyuk Lee, Yeonwoo Choi, Donggyun Kim†, Seunghoon Hong†
Under Review
Flow Map Language Models: One-step Language Modeling via Continuous Denoising
Chanhyuk Lee, Jaehoon Yoo, Manan Agarawal, Sheel Shah, Jerry Huang, Aditi Raghunathan, Seunghoon Hong, Nicholas M. Boffi†, Jinwoo Kim†
Preprint
Infinite Mask Diffusion for Few-Step Distillation
Jaehoon Yoo, Wonjung Kim, Chanhyuk Lee, Seunghoon Hong
International Conference on Machine Learning (ICML), 2026
AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
Chanhyuk Lee, Jiho Choi, Chanryeol Lee, Donggyun Kim, Seunghoon Hong
International Conference on Learning Representations (ICLR), 2026
Universal Few-Shot Spatial Control for Diffusion Models
Kiet T. Nguyen, Chanhyuk Lee, Donggyun Kim, Donghoon Lee, Seunghoon Hong
Conference on Neural Information Processing Systems (NeurIPS), 2025
Revisiting Weight Averaging for Model Merging
Jiho Choi*, Donggyun Kim*, Chanhyuk Lee, Seunghoon Hong
Preprint
Education
M.S. in Computer Science
Korea Advanced Institue of Science and Technology (KAIST) (Mar. 2025 -)
B.S. in Computer Science and Chemistry (Double Major)
Korea Advanced Institue of Science and Technology (KAIST) (Mar. 2019 - Feb. 2025)
Gyeonggi Science High School for Gifted
(Mar. 2016 - Feb. 2019)