Hello, I'm Seohyun Lee!

I am a master’s student in the School of Computing at KAIST, advised by Prof. Hyunwoo J. Kim at the MLV Lab. Before starting my graduate studies, I interned at LG AI Research in the Vision Lab’s Deep Document Understanding Squad and the EXAONE-VL Post-Training Team. I received my undergraduate degree from Korea University, where I explored AGI and generative models through KUBIG, the Data Science & AI society, under the guidance of Prof. Sungbin Lim.

My research interests lie in reliable vision-language models for multimodal reasoning and agentic interaction. I am especially interested in post-training, inference-time reasoning, and multimodal retrieval for enabling models to reason, verify, and adapt across multimodal environments.