Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs

VLMRe-assign community