|
Wooyeol Lee I'm a senior undergraduate student majoring in computer science at Seoul National University in South Korea. Currently, I'm working as an research intern at Artificial Intelligence & Biomedical Informatics Lab advised by Prof. Kyungsu Kim. My interests span artificial intelligence, bioinformatics, and robotics, focusing on high-impact technologies beyond research. I am passionate about developing innovative solutions and committed to continuous learning in emerging fields. |
|
ResearchMy research is focused on generative AI and its applications in various domains, including synthetic data generation and macromolecular design. |
|
On the Collapse of Generative Paths: A Criterion and Correction for Diffusion Steering
*, *, , Wooyeol Lee, , , , †, † arXiv preprint, 2025 paper Ratio-of-densities steering in diffusion models breaks via Marginal Path Collapse. We introduce ACE, a provably stable correction that prevents collapse, cuts error >5x, and enables reliable molecular design. |
|
ISAC: Training-Free Instance-to-Semantic Attention Control for Improving Multi-Instance
Generation
*, Wooyeol Lee*, *, † arXiv preprint, 2025 paper Aligning guidance strategy with instance-to-semantic dynamics, where instance layouts form first then semantic details are refined, is crucial for effective multi-instance generation. |
|
TRACE: Your Diffusion Model is Secretly an Instance Edge Detector
*, *, Wooyeol Lee, , †, † arXiv preprint, 2025 paper Diffusion models, capable of generating multiple instances, are secretly powerful tools for instance segmentation. |
|
Website template from Jon Barron's website. I appreciate his work in creating such a clean and simple design. |