| video-question-answering-on-activitynet-qa | FrozenBiLM | #17 | Accuracy: 43.2 |
| video-question-answering-on-activitynet-qa | FrozenBiLM (0-shot) | #30 | Accuracy: 25.9 |
| video-question-answering-on-how2qa | FrozenBiLM | #2 | Accuracy: 86.7 |
| video-question-answering-on-how2qa | FrozenBiLM (0-shot) | #6 | Accuracy: 58.4 |
| video-question-answering-on-ivqa | FrozenBiLM | #2 | Accuracy: 39.6 |
| video-question-answering-on-ivqa | FrozenBiLM (0-shot) | #4 | Accuracy: 26.8 |
| video-question-answering-on-msrvtt-qa | FrozenBiLM | #6 | Accuracy: 47.0 |
| video-question-answering-on-msrvtt-qa | FrozenBiLM (0-shot) | #13 | Accuracy: 16.7 |
| video-question-answering-on-tvqa | FrozenBiLM | #2 | Accuracy: 82 |
| visual-question-answering-on-msrvtt-qa-2 | FrozenBiLM | #1 | Accuracy: 0.470 |
| zero-shot-video-question-answer-on-egoschema-1 | FrozenBiLM | #27 | Accuracy: 26.9 |
| zero-shot-video-question-answer-on-tvqa | FrozenBiLM (with speech) | #1 | Accuracy: 59.7 |
| zero-shot-video-question-answer-on-tvqa | FrozenBILM (no speech) | #9 | Accuracy: 29.7 |
| zeroshot-video-question-answer-on-activitynet | FrozenBiLM | #27 | Accuracy: 24.7 |
| zeroshot-video-question-answer-on-msvd-qa | FrozenBiLM | #28 | Accuracy: 33.8 |
| zeroshot-video-question-answer-on-tgif-qa | FrozenBiLM | #13 | Accuracy: 41.9 |