Cross-Modal Retrieval on CC152K

Metric: Image-to-text R@1 (higher is better)

LeaderboardDataset
Loading chart...