Emotion Recognition on MELD (test)
70.12Weighted F1VowelPrompt
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| VowelPromptBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT2026.02 | 70.12 | — | — | — | |
| VowelPromptBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT & GRPO2026.02 | 69.79 | — | — | — | |
| VowelPromptBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT2026.02 | 69.61 | — | — | — | |
| VowelPromptBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT & GRPO2026.02 | 68.98 | — | — | — | |
| InstructERCBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT2026.02 | 68.15 | — | — | — | |
| SpeechCueLLMBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT2026.02 | 68.02 | — | — | — | |
| SALMONNBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT2026.02 | 67.96 | — | — | — | |
| SpeechCueLLMBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT & GRPO2026.02 | 67.96 | — | — | — | |
| BaselineBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT2026.02 | 67.9 | — | — | — | |
| DF-GCNnumber of runs=102026.03 | 67.6 | — | — | 67.4 | |
| BaselineBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT2026.02 | 67.44 | — | — | — | |
| InstructERCBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT & GRPO2026.02 | 67.35 | — | — | — | |
| InstructERCBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT2026.02 | 67.25 | — | — | — | |
| SALMONNBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT2026.02 | 67.25 | — | — | — | |
| SpeechCueLLMBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT & GRPO2026.02 | 67.1 | — | — | — | |
| SALMONNBackbone=LLaMA-4-Scout-17B-16E-Instruct, Training Regime=SFT & GRPO2026.02 | 67.1 | — | — | — | |
| SpeechCueLLMBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT2026.02 | 67.07 | — | — | — | |
| InstructERCBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT & GRPO2026.02 | 66.96 | — | — | — | |
| SALMONNBackbone=LLaMA-3-8B-Instruct, Training Regime=SFT & GRPO2026.02 | 66.85 | — | — | — | |
| M3Netnumber of runs=102026.03 | 65.8 | — | — | 63.9 | |
| AdaIGNnumber of runs=102026.03 | 65.8 | — | — | 65.3 | |
| CBERLnumber of runs=102026.03 | 65.5 | — | — | 64.4 | |
| DER-GCNnumber of runs=102026.03 | 65.5 | — | — | 65.2 | |
| MGLRAnumber of runs=102026.03 | 64.9 | — | — | 65.4 | |
| EmoBERTanumber of runs=102026.03 | 63.3 | — | — | 64.1 | |
| CoMPMnumber of runs=102026.03 | 63 | — | — | 64.3 | |
| MM-DFN + SPCLBase Architecture=MM-DFN, Modality=TAV, Plug-in=SPCL2026.05 | 59.17 | — | — | 61.91 | |
| MMGCN + SPCLBase Architecture=MMGCN, Modality=TAV, Plug-in=SPCL2026.05 | 59.11 | — | — | 61.32 | |
| MM-DFN + SPCLBase Architecture=MM-DFN, Modality=TA, Plug-in=SPCL2026.05 | 59.11 | — | — | 62.31 | |
| MMGCN + SPCLBase Architecture=MMGCN, Modality=TA, Plug-in=SPCL2026.05 | 58.93 | — | — | 61.65 | |
| MM-DFN + SPCLBase Architecture=MM-DFN, Modality=TV, Plug-in=SPCL2026.05 | 58.91 | — | — | 61.94 | |
| MM-DFN + OPMBase Architecture=MM-DFN, Modality=TAV, Plug-in=OPM2026.05 | 58.75 | — | — | 61.42 | |
| MMGCN + FAGMBase Architecture=MMGCN, Modality=TAV, Plug-in=FAGM2026.05 | 58.48 | — | — | 61.15 | |
| AffectVerseAudio=true, Video=true, Text=true2026.05 | 58.47 | — | — | — | |
| MMGCNnumber of runs=102026.03 | 58.4 | — | — | 58.3 | |
| AffectVerseAudio=true, Video=false, Text=true2026.05 | 58.32 | — | — | — | |
| MMGCN + SPCLBase Architecture=MMGCN, Modality=TV, Plug-in=SPCL2026.05 | 58.14 | — | — | 60.64 | |
| BiDDIN + SPCLBase Architecture=BiDDIN, Modality=TV, Plug-in=SPCL2026.05 | 58.1 | — | — | 61 | |
| BiDDIN + SPCLBase Architecture=BiDDIN, Modality=TA, Plug-in=SPCL2026.05 | 58.08 | — | — | 61.22 | |
| DialogueGCN + SPCLBase Architecture=DialogueGCN, Modality=TA, Plug-in=SPCL2026.05 | 58.04 | — | — | 60.84 | |
| QMNNnumber of runs=102026.03 | 58 | — | — | 60.8 | |
| DialogueGCN + SPCLBase Architecture=DialogueGCN, Modality=TAV, Plug-in=SPCL2026.05 | 57.87 | — | — | 60.77 | |
| DialogueGCN + OGM-GEBase Architecture=DialogueGCN, Modality=TAV, Plug-in=OGM-GE2026.05 | 57.73 | — | — | 57.36 | |
| MMGCN (Baseline)Base Architecture=MMGCN, Modality=TAV, Plug-in=None2026.05 | 57.71 | — | — | 59.95 | |
| AffectGPTAudio=false, Video=true, Text=true2026.05 | 57.63 | — | — | — | |
| BiDDIN + SPCLBase Architecture=BiDDIN, Modality=TAV, Plug-in=SPCL2026.05 | 57.6 | — | — | 60.86 | |
| MMGCN + OGM-GEBase Architecture=MMGCN, Modality=TAV, Plug-in=OGM-GE2026.05 | 57.59 | — | — | 59.92 | |
| MM-DFN + FAGMBase Architecture=MM-DFN, Modality=TAV, Plug-in=FAGM2026.05 | 57.55 | — | — | 60.8 | |
| MM-DFN (Baseline)Base Architecture=MM-DFN, Modality=TAV, Plug-in=None2026.05 | 57.52 | — | — | 59.9 | |
| BiDDIN + FAGMBase Architecture=BiDDIN, Modality=TAV, Plug-in=FAGM2026.05 | 57.47 | — | — | 59.18 | |
| AffectVerseAudio=false, Video=true, Text=true2026.05 | 57.38 | — | — | — | |
| MMGCN + RNA lossBase Architecture=MMGCN, Modality=TAV, Plug-in=RNA loss2026.05 | 56.94 | — | — | 58.62 | |
| DialogueGCN + RNA lossBase Architecture=DialogueGCN, Modality=TAV, Plug-in=RNA loss2026.05 | 56.65 | — | — | 58.47 | |
| AffectGPTAudio=true, Video=false, Text=true2026.05 | 56.63 | — | — | — | |
| MM-DFN + OGM-GEBase Architecture=MM-DFN, Modality=TAV, Plug-in=OGM-GE2026.05 | 56.53 | — | — | 58.39 | |
| BiDDIN (Baseline)Base Architecture=BiDDIN, Modality=TAV, Plug-in=None2026.05 | 56.41 | — | — | 58.54 | |
| DialogueGCN + SPCLBase Architecture=DialogueGCN, Modality=TV, Plug-in=SPCL2026.05 | 56.18 | — | — | 58.61 | |
| MM-DFN + RNA lossBase Architecture=MM-DFN, Modality=TAV, Plug-in=RNA loss2026.05 | 56.02 | — | — | 58.2 | |
| MMGCN + OPMBase Architecture=MMGCN, Modality=TAV, Plug-in=OPM2026.05 | 55.78 | — | — | 57.24 | |
| AffectGPTAudio=true, Video=true, Text=true2026.05 | 55.65 | — | — | — | |
| Qwen2.5-OmniAudio=true, Video=true, Text=true2026.05 | 55.29 | — | — | — | |
| BiDDIN + OGM-GEBase Architecture=BiDDIN, Modality=TAV, Plug-in=OGM-GE2026.05 | 55.27 | — | — | 53.41 | |
| Qwen2.5-OmniAudio=true, Video=false, Text=true2026.05 | 54.64 | — | — | — | |
| DialogueGCN + FAGMBase Architecture=DialogueGCN, Modality=TAV, Plug-in=FAGM2026.05 | 54.61 | — | — | 58.96 | |
| Qwen2.5-OmniAudio=false, Video=true, Text=true2026.05 | 54.48 | — | — | — | |
| DialogueGCN + OPMBase Architecture=DialogueGCN, Modality=TAV, Plug-in=OPM2026.05 | 54.47 | — | — | 57.12 | |
| BiDDIN + OPMBase Architecture=BiDDIN, Modality=TAV, Plug-in=OPM2026.05 | 53.87 | — | — | 57.62 | |
| DialogueGCN (Baseline)Base Architecture=DialogueGCN, Modality=TAV, Plug-in=None2026.05 | 53.11 | — | — | 55.08 | |
| BiDDIN + RNA lossBase Architecture=BiDDIN, Modality=TAV, Plug-in=RNA loss2026.05 | 52.18 | — | — | 49.16 | |
| mPLUG-OwlAudio=false, Video=true, Text=true2026.05 | 49.11 | — | — | — | |
| Qwen-AudioAudio=true, Video=false, Text=true2026.05 | 49.09 | — | — | — | |
| Emotion-LLaMAAudio=true, Video=true, Text=true2026.05 | 46.76 | — | — | — | |
| SALMONNAudio=true, Video=false, Text=true2026.05 | 45.62 | — | — | — | |
| Chat-UniViAudio=false, Video=true, Text=true2026.05 | 45.61 | — | — | — | |
| MMGCN + SPCLBase Architecture=MMGCN, Modality=AV, Plug-in=SPCL2026.05 | 43.79 | — | — | 49.1 | |
| MM-DFN + SPCLBase Architecture=MM-DFN, Modality=AV, Plug-in=SPCL2026.05 | 43.32 | — | — | 48.59 | |
| LLaMA-VIDAudio=false, Video=true, Text=true2026.05 | 42.75 | — | — | — | |
| BiDDIN + SPCLBase Architecture=BiDDIN, Modality=AV, Plug-in=SPCL2026.05 | 42.3 | — | — | 48.15 | |
| DialogueGCN + SPCLBase Architecture=DialogueGCN, Modality=AV, Plug-in=SPCL2026.05 | 42.28 | — | — | 46.64 | |
| VideoChatAudio=false, Video=true, Text=true2026.05 | 41.11 | — | — | — | |
| PandaGPTAudio=false, Video=true, Text=true2026.05 | 38.33 | — | — | — | |
| PandaGPTAudio=true, Video=true, Text=true2026.05 | 37.88 | — | — | — | |
| Video-ChatGPTAudio=false, Video=true, Text=true2026.05 | 37.33 | — | — | — | |
| VideoChat2Audio=false, Video=true, Text=true2026.05 | 36.64 | — | — | — | |
| PandaGPTAudio=true, Video=false, Text=true2026.05 | 31.91 | — | — | — | |
| Video-LLaVAAudio=false, Video=true, Text=true2026.05 | 30.73 | — | — | — | |
| OneLLMAudio=true, Video=false, Text=true2026.05 | 28.32 | — | — | — | |
| SECapAudio=true, Video=false, Text=true2026.05 | 25.56 | — | — | — | |
| OtterAudio=false, Video=true, Text=true2026.05 | 22.57 | — | — | — | |
| Audio-Flamingo-32026.06 | — | — | — | 56.71 | |
| BERT+MTLBase=BERT2020.10 | — | — | 61.9 | — | |
| BLSP-Emo2026.06 | — | — | — | 57.29 | |
| CNNBase=GloVe2020.10 | — | 64.25 | 55.02 | — | |
| CoMEvaluation Protocol=Average of 3 runs2021.08 | — | 71.67 | 64.9 | — | |
| CoMPMEvaluation Protocol=Average of 3 runs2021.08 | — | 73.08 | 66.52 | — | |
| CoMPM(f)Evaluation Protocol=Average of 3 runs2021.08 | — | 73.04 | 65.77 | — | |
| CoMPM(k)Evaluation Protocol=Average of 3 runs2021.08 | — | 72.67 | 66.22 | — | |
| CoMPM(s)Evaluation Protocol=Average of 3 runs2021.08 | — | 71.97 | 65.12 | — | |
| ConGCNBase=GloVe2020.10 | — | — | 57.4 | — | |
| COSMICBase=RoBERTa2020.10 | — | 73.2 | 65.21 | — |