I am a Ph.D candidate at Korea Advanced Institute of Science and Technology (KAIST) advised by Professor Jaegul Choo.
My research focuses on responsible and reliable AI, with a particular emphasis on bias mitigation. I have explored debiasing tasks in classification models and recently expanded my interests to address societal bias in models trained with large-scale datasets, especially diffusion-based text-to-image (T2I) generative models.
jeonghoon_park [at] kaist.ac.kr
Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea
M.S in Artifical Intelligence, 2020-2022
KAIST, Korea
Advisor: Prof. Jaegul Choo
B.S in Computer Science, 2015-2020
Korea University, Korea
Disentangling Subject-Irrelevant Elements against Subject in Personalized Text-to-Image Diffusion via Filtered Self-distillation
Enhancing Intrinsic Features for Debiasing via Investigating Class-Discerning Common Attributes in Bias-Contrastive Pair
Training auxiliary prototypical classifiers for explainable anomaly detection in medical image segmentation
Deep Edge-Aware Interactive Colorization against Color-Bleeding Effects
Natural Attribute-based Shift Detection
Evaluation of Out-of-Distribution Detection Performance of Self-Supervised Learning in a Controllable Environment