Cross-Modal Retrieval on CC152K

Metric: Image-to-text R@10 (higher is better)

LeaderboardDataset
Loading chart...