Vision-Language Models Struggle to Align Entities across Modalities

VLM
Re-assign community