LLM meets Vision-Language Models for Zero-Shot One-Class Classification

VLMMLLMRe-assign community