Multimodal Sentiment Analysis on CMU-MOSI Unaligned (test)
39.3Accuracy (7-Class)Full modality (ours)
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Full modality (ours)Alignment=Unaligned, Modality=Full modality2023.12 | 39.3 | 82.3 | 82.1 | 0.861 | 0.69 | |
| MulTalignment=Unaligned2019.06 | 39.1 | 81.1 | 81 | 0.889 | 0.686 | |
| OursAlignment=Unaligned, Modality=Language and vision with reconstructed audio2023.12 | 38.7 | 81.9 | 81.7 | 0.844 | 0.687 | |
| Language and vision (ours)Alignment=Unaligned, Modality=Language and vision2023.12 | 37.5 | 80.9 | 80.7 | 0.838 | 0.681 | |
| Language only (ours)Alignment=Unaligned, Modality=Language only2023.12 | 36 | 77.5 | 77.2 | 0.742 | 0.606 | |
| MoNIG2021.11 | 35.8 | 79.3 | 79.3 | 0.972 | 0.664 | |
| MoNIGmodality=pseudo2021.11 | 34.7 | 79.1 | 79.1 | 0.958 | 0.669 | |
| Vision only (ours)Alignment=Unaligned, Modality=Vision only2023.12 | 34 | 76.1 | 75.9 | 0.717 | 0.401 | |
| LF-LSTMalignment=Unaligned, fusion=Late Fusion2019.06 | 33.7 | 77.6 | 77.8 | 0.988 | 0.624 | |
| LF-LSTM2021.11 | 33.7 | 77.6 | 77.8 | 0.988 | 0.624 | |
| Evidential2021.11 | 32.9 | 78.7 | 78.7 | 0.988 | 0.651 | |
| CTC + MCTNalignment=Unaligned, alignment_strategy=CTC2019.06 | 32.7 | 75.9 | 76.4 | 0.991 | 0.613 | |
| CTC + MCTN2021.11 | 32.7 | 75.9 | 76.4 | 0.991 | 0.613 | |
| CTC + MCTNAlignment=Unaligned, Modality=Full modality2023.12 | 32.7 | 75.9 | 76.4 | 0.991 | 0.613 | |
| Gaussian2021.11 | 32.4 | 77.6 | 77.5 | 1.005 | 0.634 | |
| CTC + RAVENalignment=Unaligned, alignment_strategy=CTC2019.06 | 31.7 | 72.7 | 73.1 | 1.076 | 0.544 | |
| CTC + RAVEN2021.11 | 31.7 | 72.7 | 73.1 | 1.076 | 0.544 | |
| CTC + RAVENAlignment=Unaligned, Modality=Full modality2023.12 | 31.7 | 72.7 | 73.1 | 1.076 | 0.544 | |
| CTC + EF-LSTMalignment=Unaligned, alignment_strategy=CTC2019.06 | 31 | 73.6 | 74.5 | 1.078 | 0.542 | |
| CTC + EF-LSTM2021.11 | 31 | 73.6 | 74.5 | 1.078 | 0.542 |