Multimodal Intent Recognition On Mintrec
평가 지표
Accuracy (20 classes)
Accuracy (Binary)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Accuracy (20 classes) | Accuracy (Binary) | Paper Title | Repository |
|---|---|---|---|---|
| MulT (Text + Audio + Video) | 72.52 | 89.19 | MIntRec: A New Dataset for Multimodal Intent Recognition | |
| SPECTRA | 73.48 | - | Speech-Text Dialog Pre-training for Spoken Dialog Understanding with Explicit Cross-Modal Alignment | |
| TCL-MAP | 73.62 | - | Token-Level Contrastive Learning with Modality-Aware Prompting for Multimodal Intent Recognition | |
| Human | 85.51 | 94.72 | MIntRec: A New Dataset for Multimodal Intent Recognition | |
| MAG-BERT (Text + Audio + Video) | 72.65 | 89.24 | MIntRec: A New Dataset for Multimodal Intent Recognition | |
| MISA (Text + Audio + Video) | 72.29 | 89.21 | MIntRec: A New Dataset for Multimodal Intent Recognition |
0 of 6 row(s) selected.