| video-captioning-on-msr-vtt-1 | COSA | #4 | CIDEr: 74.7BLEU-4: 53.7 |
| video-captioning-on-msvd-1 | COSA | #3 | CIDEr: 178.5BLEU-4: 76.5 |
| video-captioning-on-vatex-1 | COSA | #3 | BLEU-4: 43.7CIDEr: 96.5 |
| video-captioning-on-youcook2 | COSA | #7 | BLEU-4: 10.1CIDEr: 1.31 |
| video-question-answering-on-activitynet-qa | COSA | #2 | Accuracy: 49.9 |
| video-question-answering-on-msrvtt-qa | COSA | #3 | Accuracy: 49.2 |
| video-retrieval-on-activitynet | COSA | #5 | text-to-video R@1: 67.3 |
| video-retrieval-on-didemo | COSA | #4 | text-to-video R@1: 70.5 |
| video-retrieval-on-lsmdc | COSA | #4 | text-to-video R@1: 39.4 |
| video-retrieval-on-msr-vtt | COSA | #7 | text-to-video R@1: 57.9 |
| visual-question-answering-on-msvd-qa-1 | COSA | #5 | Accuracy: 0.60 |