Seoul, South Korea

Hyeonseo Yu.

I am an M.S. student at KAIST AI, working with Prof. Seungryong Kim in CVLAB.

My research focuses on 3D/4D reconstruction, multimodal learning, and 3D feature representation. My long-term goal is to bridge 3D/4D perception and multimodal reasoning to enable intelligent agents to understand and interact with the physical world, with applications in robotics and embodied AI.

News.

Sep 2026
Imagine3D-LLM has been accepted to NeurIPS 2026!
Sep 2026
I joined KAIST CVLAB as an M.S. student in Artificial Intelligence.
Feb 2026
I graduated from Hanyang University summa cum laude, ranked 1st, with a GPA of 4.49/4.50.

Publications.

Google Scholar

Ask4NV Overview

Preprint · 2026Novel-view generation · Multimodal learning

Can Learning to Answer Spatial Questions Improve Novel-View Generation?

Jahyeok Koo, Hyeonseo Yu, Seungho Jang, Jaewoo Jung, Jiwon Kang, Seokju Cho, Seungryong Kim^†

Paper (soon)Project Code

Imagine3D-LLM Overview

NeurIPS 20263D understanding · Multimodal learning

Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering

Jaewoo Jung, Hyeonseo Yu, Honggyu An, Jisang Han, Mungyeom Kim, Minkyeong Jeon, Heeseong Shin, WonJun Moon, Federico Tombari, Daniel Barath, Marc Pollefeys^†, Seungryong Kim^†, Sunghwan Hong^†

Paper Project Code

Comparison of 4D scene reconstruction methods from the C4G project

C4G Enlarge

Preprint · 20264D reconstruction

Learning Temporally-Aware Compact Gaussians for Feed-Forward 4D Reconstruction

Mungyeom Kim^*, Minkyeong Jeon^*, Honggyu An^*, Jaewoo Jung, Hyunah Ko, Jisang Han, Hyeonseo Yu, Donghwan Shin, Sunghwan Hong, Takuya Narihira, Kazumi Fukuda, Yuki Mitsufuji^†, Seungryong Kim^†

Paper Project Code