Advancing Compositional Awareness in CLIP with Efficient Fine-Tuning

CoGeVLMRe-assign community