Multimodal Sentiment Analysis on CMU-MOSI
89.5F1 ScoreCmIR
Evaluation results
| Method | Links | |||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CmIR2026.04 | 89.5 | 0.616 | 0.853 | 49.8 | — | — | — | — | — | — | 89.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours2026.05 | 89.23 | 0.638 | 0.864 | — | — | — | — | — | — | — | 89.31 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ITHP2026.05 | 88.6 | 0.643 | 0.852 | — | — | — | — | — | — | — | 88.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ITHPVenue=ICLR 242026.04 | 88.5 | 0.663 | 0.856 | 47.7 | — | — | — | — | — | — | 88.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Multimodal BoostingVenue=TMM 242026.04 | 88.4 | 0.634 | 0.855 | 49.1 | — | — | — | — | — | — | 88.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GSConVenue=TIP 252026.04 | 88 | 0.696 | 0.832 | 45.7 | — | — | — | — | — | — | 88.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| C-MIBVenue=TMM 232026.04 | 87.8 | 0.662 | 0.835 | 47.7 | — | — | — | — | — | — | 87.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| QA-MoE2026.04 | 87.6 | 0.579 | 0.817 | 53.6 | — | — | — | — | — | — | 88.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| QA-MoEBackbone=BERT-base-uncased2026.04 | 87.2 | 0.583 | 0.816 | 53.3 | — | — | — | — | — | — | 87.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Diffusion BridgeVenue=CVPR 252026.04 | 86.8 | 0.649 | 0.839 | 47.3 | — | — | — | — | — | — | 86.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMA2026.04 | 86.4 | 0.693 | 0.803 | 46.9 | — | — | — | — | — | — | 86.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TLRAModality Setting=Full-modality2026.05 | 86.24 | — | — | — | — | — | — | — | 86.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EBMC2026.04 | 86.2 | 0.694 | 0.833 | 50.34 | — | — | — | — | — | — | 86.26 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAGvariant=d2026.05 | 86 | 0.69 | 0.831 | — | — | — | — | — | — | — | 86.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMIMvariant=d2026.05 | 85.9 | 0.649 | 0.829 | — | — | — | — | — | — | — | 85.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MECAMModality Setting=Full-modality2026.05 | 85.8 | — | — | — | — | — | — | — | 85.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniMSE2026.05 | 85.8 | 0.691 | 0.809 | — | — | — | — | — | — | — | 85.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EMOE2026.04 | 85.4 | 0.71 | — | 47.7 | — | — | — | — | — | — | 85.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EMOEBackbone=BERT-base-uncased2026.04 | 85.3 | 0.697 | — | 47.8 | — | — | — | — | — | — | 85.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MIB2026.05 | 85.3 | 0.711 | 0.798 | — | — | — | — | — | — | — | 85.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ConFEDEVenue=ACL 232026.04 | 85.2 | 0.728 | 0.784 | 43.3 | — | — | — | — | — | — | 85.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Semi-IIN2026.04 | 85.19 | 0.679 | 0.822 | 46.5 | — | — | — | — | — | — | 85.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| U-PNSModality Setting=Full-modality2026.05 | 85.15 | — | — | — | — | — | — | — | 85.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DLFModality Setting=Full-modality2026.05 | 85.04 | — | — | — | — | — | — | — | 85.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UEFNModality Setting=Full-modality2026.05 | 85.03 | — | — | — | — | — | — | — | 85.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ALMTModality Setting=Full-modality2026.05 | 85.01 | — | — | — | — | — | — | — | 84.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Self-MMVenue=AAAI 212026.04 | 84.8 | 0.731 | 0.785 | 45.8 | — | — | — | — | — | — | 84.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EMOEVenue=CVPR 252026.04 | 84.8 | 0.723 | 0.79 | 45.2 | — | — | — | — | — | — | 84.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MFON2026.04 | 84.75 | 0.725 | 0.797 | 44.9 | — | — | — | — | — | — | 84.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FDMERVenue=ACM MM 222026.04 | 84.7 | 0.724 | 0.788 | 44.1 | — | — | — | — | — | — | 84.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LNLNModality Setting=Full-modality2026.05 | 84.61 | — | — | — | — | — | — | — | 84.25 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| M-BERTModality Setting=Full-modality2026.05 | 84.61 | — | — | — | — | — | — | — | 84.43 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMIM2026.04 | 84.5 | 0.717 | — | 45.8 | — | — | — | — | — | — | 84.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SuCIVenue=AAAI 252026.04 | 84.5 | — | — | 42.2 | — | — | — | — | — | — | 84.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DEVA2026.04 | 84.48 | 0.73 | 0.787 | 46.32 | — | — | — | — | — | — | 84.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SelfMMsource_notation=♢2026.04 | 84.42 | 0.712 | 0.795 | 45.79 | — | — | — | — | — | — | 82.54 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Self-MMvariant=b2026.05 | 84.4 | 0.713 | 0.798 | — | — | — | — | — | — | — | 84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ConKIsource_notation=♢2026.04 | 84.33 | 0.681 | 0.816 | 48.43 | — | — | — | — | — | — | 84.37 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BBFN2026.05 | 84.3 | 0.776 | 0.755 | — | — | — | — | — | — | — | 84.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EUARreproduced=true2026.04 | 84.22 | 0.711 | 0.789 | 48.18 | — | — | — | — | — | — | 83.24 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MISA2026.04 | 84.2 | 0.754 | 0.761 | 41.8 | — | — | — | — | — | — | 84.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ConFEDE2026.04 | 84.13 | 0.784 | 0.742 | 42.27 | — | — | — | — | — | — | 84.17 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAGvariant=b2026.05 | 84.1 | 0.712 | 0.796 | — | — | — | — | — | — | — | 84.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMIMvariant=b2026.05 | 84 | 0.7 | 0.8 | — | — | — | — | — | — | — | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MISABackbone=BERT-base-uncased2026.04 | 83.9 | 0.742 | 0.761 | 43.6 | — | — | — | — | — | — | 83.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GLoMosource_notation=♡2026.04 | 83.9 | 0.718 | 0.782 | 48.3 | — | — | — | — | — | — | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CLGSI2026.04 | 83.63 | 0.703 | 0.79 | 47.96 | — | — | — | — | — | — | 83.97 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMIMBackbone=BERT-base-uncased2026.04 | 83.4 | 0.777 | — | 45.9 | — | — | — | — | — | — | 83.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DMDModality Setting=Full-modality2026.05 | 83.29 | — | — | — | — | — | — | — | 82.23 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(0.0, –, 0.0), inter=AT2026.07 | 81.08 | — | — | — | — | — | — | — | 81.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(–, 0.0, 0.0), inter=VT2026.07 | 80.8 | — | — | — | — | — | — | — | 80.85 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(0.0, 0.0, 0.0), inter=AVT2026.07 | 80.76 | — | — | — | — | — | — | — | 80.82 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MulTBackbone=BERT-base-uncased2026.04 | 80.3 | 0.933 | 0.711 | 33.2 | — | — | — | — | — | — | 80.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MulT2026.04 | 80.1 | 0.936 | 0.711 | 35.1 | — | — | — | — | — | — | 80 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MuLTsource_notation=♡2026.04 | 79.46 | 0.892 | 0.667 | 36.74 | — | — | — | — | — | — | 79.51 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(–, 0.0, 0.0), inter=VT2026.07 | 79.25 | — | — | — | — | — | — | — | 79.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(0.0, –, 0.0), inter=AT2026.07 | 79.23 | — | — | — | — | — | — | — | 79.24 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LMFBackbone=BERT-base-uncased2026.04 | 79.1 | 0.963 | 0.695 | 31.1 | — | — | — | — | — | — | 79.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(0.0, 0.0, 0.0), inter=AVT2026.07 | 79.02 | — | — | — | — | — | — | — | 79.07 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TFN2026.04 | 78.9 | 0.953 | 0.698 | 31.9 | — | — | — | — | — | — | 78.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LMF2026.04 | 78.7 | 0.931 | 0.695 | 36.9 | — | — | — | — | — | — | 78.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(0.5, –, 0.1), inter=AT2026.07 | 78.63 | — | — | — | — | — | — | — | 78.78 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(–, 0.5, 0.1), inter=VT2026.07 | 78.55 | — | — | — | — | — | — | — | 78.57 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(–, –, 0.1), inter=T2026.07 | 78.41 | — | — | — | — | — | — | — | 78.45 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(0.5, 0.3, 0.1), inter=AVT2026.07 | 77.6 | — | — | — | — | — | — | — | 77.67 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(–, 0.5, 0.1), inter=VT2026.07 | 77.48 | — | — | — | — | — | — | — | 77.49 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(–, –, 0.1), inter=T2026.07 | 77.15 | — | — | — | — | — | — | — | 77.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(0.5, 0.3, 0.1), inter=AVT2026.07 | 77.01 | — | — | — | — | — | — | — | 76.97 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AFFECTGPT#PARAMS=7B2026.03 | 76.83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(–, –, 0.1), inter=T2026.07 | 76.83 | — | — | — | — | — | — | — | 76.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TFNBackbone=BERT-base-uncased2026.04 | 76.6 | 0.995 | 0.698 | 35.3 | — | — | — | — | — | — | 76.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(0.0, –, 0.0), inter=AT2026.07 | 76.53 | — | — | — | — | — | — | — | 76.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(0.5, –, 0.1), inter=AT2026.07 | 76.16 | — | — | — | — | — | — | — | 76.09 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(–, 0.0, 0.0), inter=VT2026.07 | 75.82 | — | — | — | — | — | — | — | 75.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(–, 0.5, 0.1), inter=VT2026.07 | 74.63 | — | — | — | — | — | — | — | 74.61 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HyDRA#PARAMS=0.5B2026.03 | 72.59 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MPLUG-OWL#PARAMS=7B2026.03 | 72.26 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BASELINE#PARAMS=7B2026.03 | 71.93 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(0.5, –, 0.1), inter=AT2026.07 | 71.61 | — | — | — | — | — | — | — | 71.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(0.0, 0.0, 0.0), inter=AVT2026.07 | 71.34 | — | — | — | — | — | — | — | 71.34 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(0.5, 0.3, 0.1), inter=AVT2026.07 | 70.89 | — | — | — | — | — | — | — | 70.82 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(0.1, –, 0.5), inter=AT2026.07 | 69.95 | — | — | — | — | — | — | — | 69.88 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(–, 0.1, 0.5), inter=VT2026.07 | 68.61 | — | — | — | — | — | — | — | 68.54 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(–, –, 0.5), inter=T2026.07 | 68.46 | — | — | — | — | — | — | — | 68.48 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GIMLintra ratio (ra, rv, rt)=(0.1, 0.3, 0.5), inter=AVT2026.07 | 67.55 | — | — | — | — | — | — | — | 67.55 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VIDEOCHAT2#PARAMS=7B2026.03 | 67.23 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(0.1, –, 0.5), inter=AT2026.07 | 66.95 | — | — | — | — | — | — | — | 66.94 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(–, –, 0.5), inter=T2026.07 | 66.71 | — | — | — | — | — | — | — | 66.64 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(–, –, 0.5), inter=T2026.07 | 66.33 | — | — | — | — | — | — | — | 66.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(0.1, –, 0.5), inter=AT2026.07 | 66.08 | — | — | — | — | — | — | — | 66.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(–, 0.1, 0.5), inter=VT2026.07 | 66.08 | — | — | — | — | — | — | — | 66.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TMDCintra ratio (ra, rv, rt)=(0.1, 0.3, 0.5), inter=AVT2026.07 | 65.46 | — | — | — | — | — | — | — | 65.36 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VIDEOCHAT#PARAMS=7B2026.03 | 65.09 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LLAMA-VID#PARAMS=7B2026.03 | 62.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(0.1, 0.3, 0.5), inter=AVT2026.07 | 62.62 | — | — | — | — | — | — | — | 62.54 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| T2DRintra ratio (ra, rv, rt)=(–, 0.1, 0.5), inter=VT2026.07 | 62.49 | — | — | — | — | — | — | — | 62.45 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PANDAGPT#PARAMS=7B2026.03 | 61.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VIDEO-CHATGPT#PARAMS=7B2026.03 | 57.77 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VIDEO-LLAVA#PARAMS=7B2026.03 | 57.62 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CHAT-UNIVI#PARAMS=7B2026.03 | 57.62 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |