1 paper
Minghong Zhong, Guoshuai Zou, Kanghao Chen +2
Recently impressive performance has been achieved in Concept Bottleneck Models (CBM) by utilizing the image-text alignment learned by a large pre-trained vision-language model (i.e…