I am a first-year PhD student at KAIST School of Computing, VLLAB, supervised by Professor Seunghoon Hong. I received my MSc and B.S. degrees
(Magna Cum Laude) from KAIST School of Computing.
My current research interest is few-step generative modelling.
I have worked on spatial reasoning with VLMs, and on enabling versatile controllability for T2I diffusion models without heavy task-specific training.
Email: kietngt00@kaist.ac.kr
GitHub / LinkedIn
Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models (NeurIPS 2026) [paper][code]
Kiet T. Nguyen, Hanbo Shim, Jinwoo Kim†, Seunghoon Hong†
Universal Few-Shot Spatial Control for Diffusion Models (NeurIPS 2025) [paper][code]
Kiet T. Nguyen, Chanhyuk Lee, Donggyun Kim, Dong Hoon Lee, Seunghoon Hong
2023 - 2024
KAIST VLLAB, Research Intern
Development of Short-term precipitation prediction technology using Artificial Intelligence
(인공지능 기반 초단기 강수예측 기술 개발)
2025
Outstanding TA, School of Computing, KAIST
2024
Excellent Award, Undergraduate Research Program, KAIST
Project Title: Temporal Consistency Modeling of Radar Maps for Weather Forecasting.
(Mentor: Jaehoon Yoo, Advisor: Seunghoon Hong)
2023
Qualcomm - KAIST Innovation Award
Fall 2024/2025
TA, Introduction to Deep Learning (CS371), KAIST