Hyelin Nam

I'm a master student at KAIST, supervised by Professor Jong Chul Ye. Before starting my Master's degree, I was supervised by Professor Jaegul Choo.

My goal is to advance generative models beyond creation, enabling them to understand and make decisions in the real world. To achieve this, I have focused on improving diffusion models’ adaptability to specific tasks and conditions. I am also interested in extracting robust representations from them and expanding their application to diverse modalities like video, 3D and motion.

**I am actively looking for PhD openings starting in Fall 2025**

Email  /  CV  /  Scholar  /  Github

profile photo
CFG++: Manifold-constrained Classifier Free Guidance For Diffusion Models
Hyungjin Chung*, Jeongsol Kim*, Geon Yeong Park*, Hyelin Nam*, Jong Chul Ye
arXiv, 2024
project page / arXiv / code

A simple fix to CFG that enables lower guidance scales, improves sample quality and invertibility.

Contrastive Denoising Score for Text-guided Latent Diffusion Image Editing
Hyelin Nam, Gihyun Kwon, Geon Yeong Park, Jong Chul Ye
CVPR, 2024
project page / arXiv / code

Ensure structural correspondence by leveraging diffusion features during the score distillation process.

HairFIT: Pose-invariant Hairstyle Transfer via Flow-based Hair Alignment and Semantic-region-aware Inpainting
Chaeyeon Chung*, Taewoo Kim*, Hyelin Nam*, Seunghwan Choi, Gyojung Gu, Sunghyun Park, Jaegul Choo
BMVC, Oral Presentation, 2022
Best Paper Award, Korean Artificial Intelligence Association, 2021
arXiv