Dialogue Emotion Detection on DailyDialog
0.6167Micro F1 (- neutral)RoBERTa + FSA
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RoBERTa + FSABase Model=RoBERTa, Grayscale Label Type=future-self-adjust2022.06 | 0.6167 | 0.5584 | |
| COMPM + SBase Model=COMPM, Grayscale Label Type=self2022.06 | 0.611 | 0.5288 | |
| Psychological + SABase Model=Psychological, Grayscale Label Type=self-adjust2022.06 | 0.6079 | 0.55 | |
| RoBERTa + SBase Model=RoBERTa, Grayscale Label Type=self2022.06 | 0.6067 | 0.5389 | |
| COMPM + SABase Model=COMPM, Grayscale Label Type=self-adjust2022.06 | 0.6066 | 0.5259 | |
| PsychologicalBase Model=Psychological, Grayscale Label Type=None2022.06 | 0.6054 | 0.5238 | |
| COMPMBase Model=COMPM, Grayscale Label Type=None2022.06 | 0.6041 | 0.5246 | |
| RoBERTa + WBase Model=RoBERTa, Grayscale Label Type=word-embedding2022.06 | 0.6015 | 0.5177 | |
| Psychological + SBase Model=Psychological, Grayscale Label Type=self2022.06 | 0.5998 | 0.5363 | |
| RoBERTa + SABase Model=RoBERTa, Grayscale Label Type=self-adjust2022.06 | 0.5973 | 0.5375 | |
| DAG-ERC + SBase Model=DAG-ERC, Grayscale Label Type=self2022.06 | 0.594 | — | |
| DAG-ERC + SABase Model=DAG-ERC, Grayscale Label Type=self-adjust2022.06 | 0.5933 | — | |
| DAG-ERCBase Model=DAG-ERC, Grayscale Label Type=None2022.06 | 0.5925 | — | |
| RoBERTa + CBase Model=RoBERTa, Grayscale Label Type=category2022.06 | 0.5891 | 0.5344 | |
| RoBERTa (ours)Base Model=RoBERTa, Grayscale Label Type=None2022.06 | 0.5863 | 0.5117 | |
| COSMIC2021.06 | 0.5848 | 0.5105 | |
| TODKATAverage of ten runs=true, latent topics=true, knowledge base=ConceptNet2021.06 | 0.5847 | 0.5256 | |
| COG-BART + SBase Model=COG-BART, Grayscale Label Type=self2022.06 | 0.5715 | 0.5714 | |
| COG-BART + SABase Model=COG-BART, Grayscale Label Type=self-adjust2022.06 | 0.5657 | 0.5607 | |
| KATSBERTAverage of ten runs=true, knowledge base=SBERT2021.06 | 0.5578 | 0.5173 | |
| COG-BARTBase Model=COG-BART, Grayscale Label Type=None2022.06 | 0.5552 | 0.5399 | |
| TODKAT (-Topics)Average of ten runs=true, latent topics=false2021.06 | 0.5549 | 0.5136 | |
| KATCOMETAverage of ten runs=true, knowledge base=COMET2021.06 | 0.5462 | 0.5102 | |
| DialogueGCNEvaluation=Reproduced code2021.06 | 0.5373 | 0.4995 | |
| KET2021.06 | 0.5348 | — | |
| TODKAT (-KB)Average of ten runs=true, knowledge base=false2021.06 | 0.5344 | 0.5003 | |
| HiGRUEvaluation=Reproduced code2021.06 | 0.519 | 0.4904 |