Sangyun Lee

I’m Sangyun Lee (pronounced as “Sang-Yoon”), a second-year Ph.D. student in Electrical and Computer Engineering at Carnegie Mellon University advised by Giulia Fanti. I’m currently a research intern at NVIDIA. I obtained my Bachelor’s degree in Computer Science at Soongsil University. During my undergraduate years, I worked with Professors Jaegul Choo and Jong Chul Ye. Previously, I was a research intern at SI Analytics, Kakao Enterprise, and NAVER AI Lab.

Contact: sangyunl@andrew.cmu.edu

Research Interest

I work on deep generative modeling and its application in developing machine intelligence that surpasses human capabilities.

News

  • [Oct. 2024] One paper has been accepted to NeurIPS 2024.
  • [Jan. 2024] I will start my internship at NVIDIA this summer (Host: Arash Vahdat)

Research

Truncated Consistency Models

Sangyun Lee, Yilun Xu, Tomas Geffner, Giulia Fanti, Karsten Kreis, Arash Vahdat, Weili Nie

arxiv preprint

Abstract

Improving the Training of Rectified Flows

Sangyun Lee, Zinan Lin, Giulia Fanti

NeurIPS 2024

Abstract Code

Sequential Data Generation with Groupwise Diffusion Process

Sangyun Lee, Gayoung Lee, Hyunsu Kim, Junho Kim, Youngjung Uh

arxiv preprint, also appeared at ICML 2023 Workshop on Structured Probabilistic Inference & Generative Modeling

Abstract

Minimizing Trajectory Curvature of ODE-based Generative Models

Sangyun Lee, Beomsu Kim, Jong Chul Ye

ICML 2023

Abstract Code

Progressive Deblurring of Diffusion Models for Coarse-to-Fine Image Synthesis

Sangyun Lee, Hyungjin Chung, Jaehyeon Kim, Jong Chul Ye

NeurIPS 2022 Workshop on Score-Based Methods

Abstract Code

High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions

Sangyun Lee*, Gyojung Gu*, Sunghyun Park, Seunghwan Choi, Jaegul Choo

ECCV 2022

Abstract Code

Learning Multiple Probabilistic Degradation Generators for Unsupervised Real World Image Super Resolution

Sangyun Lee, Sewoong Ahn, Kwangjin Yoon

ECCV 2022 Workshop on Learning from Limited and Imperfect Data

Abstract

(* denotes equal contributions.)

Talk

Modulabs (2022.11.24 ~ 2022.12.08)

Patent

Sangyun Lee and Kwangjin Yoon, “Super Resolution Imaging Method Using Collaborative Learning.” Korean Patent 1024062870000, filed Dec 31, 2021, and issued June 2, 2022.