Metric: Accuracy (20 classes) (higher is better)
| # | Model↕ | Accuracy (20 classes)▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | Human | 85.51 | No | MIntRec: A New Dataset for Multimodal Intent Rec... | 2022-09-09 | Code |
| 2 | TCL-MAP | 73.62 | No | Token-Level Contrastive Learning with Modality-A... | 2023-12-22 | Code |
| 3 | SPECTRA | 73.48 | No | Speech-Text Dialog Pre-training for Spoken Dialo... | 2023-05-19 | Code |
| 4 | MAG-BERT (Text + Audio + Video) | 72.65 | No | MIntRec: A New Dataset for Multimodal Intent Rec... | 2022-09-09 | Code |
| 5 | MulT (Text + Audio + Video) | 72.52 | No | MIntRec: A New Dataset for Multimodal Intent Rec... | 2022-09-09 | Code |
| 6 | MISA (Text + Audio + Video) | 72.29 | No | MIntRec: A New Dataset for Multimodal Intent Rec... | 2022-09-09 | Code |