| cross-modal-retrieval-on-coco-2014 | Oscar | #17 | Text-to-image R@1: 57.5Text-to-image R@5: 82.8… |
| image-captioning-on-coco-captions | Oscar | #10 | BLEU-4: 41.7CIDER: 140METEOR: 30.6SPICE: 24.5 |
| image-captioning-on-nocaps-val-overall | OSCAR | #9 | CIDEr: 80.9SPICE: 11.3Pretrain (#images): 345M |
| image-retrieval-on-coco | Oscar | #6 | Recall@10: 98.3 |
| image-text-matching-on-commercialadsdataset | OSCAR | #3 | ADD(S) AUC: 87.45 |
| image-to-text-retrieval-on-coco | Oscar | #7 | Recall@10: 99.8 |
| visual-question-answering-on-vqa-v2-test-dev | Oscar | #17 | Accuracy: 73.82 |