CoV-Align: Efficient Fine-grained Cross-Modal Alignment with Cohesive Visual Semantics Priority
Open paper
Benchmark
Model
Rank
Results
cross-modal-retrieval-on-flickr30k
CoV-Align (Swin base-384 + BERT base)
–
Image-to-text R@1: 90.2
Rank counts only results with a code link.