I am a Co-founder and Chief Scientist at Familiar Labs Inc., where I lead the development of unified audio-visual AI models for expressive multilingual video generation and dubbing. Concurrently, I am a Ph.D. student in the Department of Artificial Intelligence at Yonsei University and a graduate researcher at the Visual Intelligence Lab (ViLab), advised by Prof. Youngjung Uh.

My research focuses on multimodal generative models, particularly audio-video generation, editing, and translation using flow matching and diffusion models. Specifically, I explore how to generate natural, temporally synchronized audiovisual content across languages while improving inference efficiency and visual fidelity.

Before joining Yonsei, I earned my B.S. in IT Engineering and Statistics from Sookmyung Women’s University and worked as an undergraduate research intern at the Computer Vision Lab at Korea University.

• Co-founder & Chief Scientist, Familiar Labs Inc.  —  June 2026 - Present

• AI Researcher, CINELINGO Lab  —  Jan 2026 - June 2026

• Graduate Student Researcher, ViLab, Yonsei Univ.(Advisor: Prof. Youngjung Uh)  —  Sep 2024 - Present

• Undergraduate Research Intern, Computer Vision Lab, Korea Univ.  —  Apr 2023 - May 2024

• Integrated M.S. - Ph.D. in Artificial Intelligence, Yonsei Univ.  —  Sep 2024 - Present

• B.S. in IT Engineering & Statistics, Sookmyung Women's Univ.  —  Mar 2020 - Feb 2024

• ❤️ [2026] "TextAway: Mask-Free Video Text Removal with End-to-End Text-Aware Generation" is under review.

• ❤️ [2026] "FlowBlending: Stage-Aware Multi-Model Sampling for Fast and High-Fidelity Video Generation" is arxived.

• ❤️ [2025] "Syncphony: Synchronized Audio-to-Video Generation with Diffusion Transformers" is accepted in ICLR 2026!

❤️ : First author or co-first author   🧡 : Co-working papers!