Yujin Bae

Hi! I am Yujin Bae, a second-year M.S. student at KAIST, where I am advised by Prof. Kuk-Jin Yoon in the Visual Intelligence Lab.

I am interested in building multimodal large language models that genuinely understand the 3D world. My current work centers on human action understanding, leveraging 3D scene representations to predict and reason about human motion. Building on this, my broader goal is to strengthen robot intelligence with language and multimodality, enabling robots to act reliably and intelligently in the human-centered spaces we live in.

Concretely, my interests span:

• Multi-modal LLMs — grounding language models in the 3D world
• Egocentric vision — reasoning about the world from a first-person view
• Human action anticipation — predicting what a person will do next
• Human–robot interaction — letting robots assist people proactively

Before joining KAIST, I received my B.S. at Ewha Womans University (ranked 1st in class!), and spent a semester as an exchange student at the University of Vermont.

Feel free to reach out via yujinbae@kaist.ac.kr!