Seoul National
University
University
Dept. of Electrical &
Computer Engineering
DSAIL Lab
Prof. Sungroh Yoon
오 영 탁
Yeongtak Oh
Ph.D. Candidate · ECE · SNU
oyt9306@gmail.com
oyt9306.github.io
Research Fields
Generative Models
Diffusion · 3D Generation · Style Transfer
Multi-Modal Language Models
Vision–Language · Omnimodal · Reasoning
Post-Training & Personalization
On-Policy RL · Alignment · Physical AI
Yeongtak Oh 오영탁
Ph.D. Candidate · DSAIL · Seoul National University
Expected Graduation: August 2027
Homepage
Research Vision
"Building AI that perceives, reasons, and creates in personalized, human-aligned ways — through expressive post-training of generative and multi-modal systems."
Selected Publications
(First Author Only)
arXiv '26
Contextualized Visual Personalization in Vision-Language Models
NeurIPS '25
RePIC: Reinforced Post-Training for Personalizing Multi-Modal Language Models
ECCV '24
Efficient Diffusion-Driven Corruption Editor for Test-Time Adaptation
BMVC '24
ControlDreamer: Blending Geometry and Style in Text-to-3D
IJCV '24
On Mitigating Stability-Plasticity Dilemma in CLIP-Guided Image Morphing