- Natural Language Processing
- Representation Learning
- Mechanistic Interpretability
Suhyeong Park
I am a Ph.D. student in Computer Science at Korea University’s DMIS Lab, advised by Prof. Jaewoo Kang.
My research focuses on NLP, representation learning, and mechanistic interpretability. I study what neural networks encode in their internal representations and how they use those representations to make predictions.
Recently
All news-
Co-authored SwitchMem, which learns when GUI agents should remember in text or images.
-
Released CVRR, a study of how vision-language models use latent visual reasoning.
-
Started my Ph.D. in Computer Science at Korea University’s DMIS Lab.
-
Released SaMer, which merges visual tokens while preserving evidence for retrieval.