| Benchmark | Model | Rank | Results |
|---|---|---|---|
| multimodal-intent-recognition-on-mintrec | Human | #1 | Accuracy (20 classes): 85.51Accuracy (Binary): 94.72 |
| multimodal-intent-recognition-on-mintrec | MAG-BERT (Text + Audio + Video) | #4 | Accuracy (20 classes): 72.65Accuracy (Binary): 89.24 |
| multimodal-intent-recognition-on-mintrec | MulT (Text + Audio + Video) | #5 | Accuracy (20 classes): 72.52Accuracy (Binary): 89.19 |
| multimodal-intent-recognition-on-mintrec | MISA (Text + Audio + Video) | #6 | Accuracy (20 classes): 72.29Accuracy (Binary): 89.21 |