3 papers
cs.CV2025
Comparison Reveals Commonality: Customized Image Generation through Contrastive Inversion
Minseo Kim, Minchan Kwon, Dongyeun Lee +2
The recent demand for customized image generation raises a need for techniques that effectively extract the common concept from small sets of images. Existing methods typically rel…
eess.AS2025
FairASR: Fair Audio Contrastive Learning for Automatic Speech Recognition
Jongsuk Kim, Jaemyung Yu, Minchan Kwon +1
Large-scale ASR models have achieved remarkable gains in accuracy and robustness. However, fairness issues remain largely unaddressed despite their critical importance in real-worl…
cs.LG2023
Enhancing Accuracy and Robustness through Adversarial Training in Class Incremental Continual Learning
Minchan Kwon, Kangil Kim
In real life, adversarial attack to deep learning models is a fatal security issue. However, the issue has been rarely discussed in a widely used class-incremental continual learni…