Factual Consistency Evaluation on XSumFaith (test)
42.5Pearson Correlation CoefficientDAE
Evaluation results
| Method | Links | |
|---|---|---|
| DAEType=NLI2023.05 | 42.5 | |
| QuestEvalType=QA2023.05 | 41.9 | |
| BLEURTType=Regression2023.05 | 38.7 | |
| ALIGNSCORE-baseType=Ours2023.05 | 38.2 | |
| ROUGE-1Type=Similarity Matching2023.05 | 36.1 | |
| ALIGNSCORE-largeType=Ours2023.05 | 31.1 | |
| SimCSEType=Similarity Matching2023.05 | 30.9 | |
| ROUGE-LType=Similarity Matching2023.05 | 30.6 | |
| QAFactEvalType=QA2023.05 | 30.3 | |
| BARTScoreType=Misc2023.05 | 29.3 | |
| ROUGE-2Type=Similarity Matching2023.05 | 27.6 | |
| CTCType=Misc2023.05 | 27.2 | |
| UniEvalType=Misc2023.05 | 23.9 | |
| NER-OverlapType=Similarity Matching2023.05 | 21.9 | |
| BLEUType=Similarity Matching2023.05 | 18.9 | |
| MNLIType=NLI2023.05 | 15.8 | |
| BERTScoreType=Similarity Matching2023.05 | 13 | |
| SummaC-CONVType=NLI2023.05 | 10.2 | |
| BLANCType=Misc2023.05 | 8.3 | |
| SummaC-ZSType=NLI2023.05 | 6.4 | |
| FactCCType=Misc2023.05 | 4.9 | |
| FEQAType=QA2023.05 | 1.3 |