Research Interest

My primary focus is on Generative Models, Diffusion Models, Video Editing, Image Editing. My recent works include:

• Enhanced video editing by developing a ‘motion word’ approach, enabling subject replacement across diverse body structures and increasing text alignment up to 7.66%.
• Identified and resolved a conflict among embeddings in zero-shot text-to-image (T2I) customization using orthogonal embedding, improving text alignment by 3.82-5.26% and image alignment by 2.11-12.2%.

Education

Seoul National University
Ph.D. in Computer Science, advised by Prof. Nojun Kwak.
Supported by LG Electronics Scholarship 2023-2025
Mar 2020 - Feb 2026 (Expected)
Seoul National University
B.S. in Statistics and Psychology (double-major).
Mar 2013 - Feb 2020

Employment

Meta | Research Scientist Intern
• Research on building efficient image generation model.
Aug 2025 - Jan 2026
Seattle, WA
NAVER Webtoon | Research Scientist Intern
• Research on video generation model using the motion token concept. Published ECCV'24.
Jan 2023 - Jul 2023
Seongnam, Korea

Selected Publications

Point-to-Point: Sparse Motion Guidance for Controllable Video Editing
Yeji Song, Jaehyun Lee, Mijin Koo, Nojun Kwak
https://ldynx.github.io/point-to-point/
ArXiv, 2025
Targeted Data Protection for Diffusion Model by Matching Training Trajectory
Hojun Lee, Mijin Koo, Yeji Song, Nojun Kwak
AAAI, 2026
ReFlex: Text-Guided Editing of Real Images in Rectified Flow via Mid-Step Feature Extraction and Attention Adaptation
Jimyeong Kim, Jungwon Park, Yeji Song, Nojun Kwak, Wonjong Rhee
https://wlaud1001.github.io/ReFlex/
ICCV, 2025 Highlight
Harmonizing Visual and Textual Embeddings for Zero-Shot Text-to-Image Customization
Yeji Song, Jimyeong Kim, Wonhark Park, Wonsik Shin, Wonjong Rhee, Nojun Kwak
https://ldynx.github.io/harmony-zero-t2i/
AAAI, 2025
SAVE: Protagonist Diversification with Structure Agnostic Video Editing
Yeji Song, Wonsik Shin, Junsoo Lee, Jeesoo Kim, Nojun Kwak
https://ldynx.github.io/SAVE/
ECCV, 2024
Towards Efficient Neural Scene Graphs by Learning Consistency Fields
Yeji Song, Chaerin Kong, Seoyoung Lee, Nojun Kwak, Joonseok Lee
BMVC, 2022

Patents

Apparatus and Method for Generating Object Images Based on Learning Consistency Fields. KR Patent 10-2022-0155016 (Filed)

Reviewer

• AAAI 2026
• ICCV 2025
• CVPR 2025
• ACM MM 2023 - 2025

Teaching

• TA, Introduction to Diffusion-based Generative Models for The AI Lab (Feb 2024)
• TA, Linear Algebra for Hyundai Heavy Industries. (Mar 2022 - Jun 2024)
• TA, Mathematics for Intelligent Systems (Mar 2023 - Jun 2023)