| Benchmark | Model | Rank | Results |
|---|---|---|---|
| cross-modal-retrieval-on-coco-2014 | Dual-Path (ResNet) | #30 | Text-to-image R@1: 25.3Text-to-image R@5: 53.4… |
| image-captioning-on-flickr30k-captions-test | BRNN | #2 | BLEU-4: 15.7CIDEr: 24.7METEOR: 15.3 |
| image-retrieval-on-flickr30k-1k-test | DVSA (R-CNN, AlexNet) | #15 | R@1: 15.2R@10: 50.5 |
| image-to-text-retrieval-on-coco | DVSA | #8 | Recall@10: 74.8 |