| Benchmark | Model | Rank | Results |
|---|---|---|---|
| audio-visual-question-answering-on-music-avqa | VideoLLaMA2-AV + MAD | #3 | Acc: 79.1 |
| audio-visual-question-answering-on-omnibench | VideoLLaMA2-AV + MAD | #7 | Accuracy: 36.8 |
| audio-visual-question-answering-on-worldsense | VideoLLaMA2-AV + MAD | #17 | Accuracy: 25.6 |