Hyunjin Seo
I am a Ph.D. student at Machine Learning and Intelligence Lab (MLILAB) at KAIST, under the supervision of Prof. Eunho Yang. My primary research focuses on the practical applications of geometric deep learning. This includes relation-guided multimodal understading, where I integrate large language models/vision-language models with geometric deep learning to enhance their reasoning capabilities across diverse data modalities. Additionally, I have developed data and model pruning techniques to address the computational challenges associated with training large-scale graphs, maximizing the utility of deep learning for structured data. My research also extends into the chemistry domain, where I am devising bias-mitigated framework for precise molecular conformation prediction.
