| Benchmark | Model | Rank | Results |
|---|---|---|---|
| image-classification-on-imagenet | Ours-RGB0.3 | – | Top 1 Accuracy: 63.1% |
| zero-shot-cross-modal-retrieval-on-coco-2014 | Ours-Embedding | – | Image-to-text R@1: 23.77 |
| zero-shot-cross-modal-retrieval-on-flickr30k | Ours-Embedding | – | Text-to-image R@1: 57.0 |