Vision-Language Model

Rectify representation bias in vision-language models for long-tailed recognition