Visual Question Answering On Msrvtt Qa 2
評価指標
Accuracy
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Accuracy | Paper Title | Repository |
|---|---|---|---|
| Just Ask | 0.415 | Just Ask: Learning to Answer Questions from Millions of Narrated Videos | |
| SSML | 0.35 | Noise Estimation Using Density Estimation for Self-Supervised Multimodal Learning | |
| Aurora (ours, r=64) Aurora (ours, r=64) | - | - | - |
| FrozenBiLM | 0.470 | Zero-Shot Video Question Answering via Frozen Bidirectional Language Models |
0 of 4 row(s) selected.