Words or Vision: Do Vision-Language Models Have Blind Faith in Text?

VLM
Re-assign community