Do Vision-Language Models Understand Compound Nouns?

CLIPCoGeVLMRe-assign community