Publications include: ReDi: Rectified Discrete Flow (NeurIPS 2025), Simulation-Free Training of Neural ODEs on Paired Data (NeurIPS 2024), Learning to Compose: Improving Object Centric Learning by Injecting Compositionality (ICLR 2024), Towards End-to-End Generative Modeling of Long Videos with Memory-Efficient Bidirectional Transformers (CVPR 2023), SetVAE: Learning Hierarchical Composition for Generative Modeling of Set-Structured Data (CVPR 2021). Projects: Development of Short-term precipitation prediction technology using Artificial Intelligence (2021-2024), Line-art Colorization with SPADE (2019), Reproducing PiCANet: Learning Pixel-wise Contextual Attention for Saliency Detection with Pytorch (2018). Honors: ICLR2025 Notable Reviewer, The Presidential Science Scholarship (2016-2020).
Research Experience
Research Intern at KAIST VLLab (2020), KAIST SIIT Lab (2019), NAVER WEBTOON (2018), and KAIST IVY Lab (2018).
Education
Received B.S. in Electrical Engineering and M.S. in Computer Science from KAIST. Advisor: Prof. Seunghoon Hong.
Background
PhD student at KAIST VLLAB. Research interest in improving the computational efficiency of neural networks and enhancing generative models.
Miscellany
Contact: wogns98@kaist.ac.kr. Other platforms: Google Scholar, GitHub