Factual Consistency Evaluation on QAGS-CNNDM (test)
78.1Pearson Correlation CoefficientALIGNSCORE-large
Evaluation results
| Method | Links | |
|---|---|---|
| ALIGNSCORE-largeType=Ours2023.05 | 78.1 | |
| ALIGNSCORE-baseType=Ours2023.05 | 72.3 | |
| BARTScoreType=Misc2023.05 | 71.5 | |
| QAFactEvalType=QA2023.05 | 68.4 | |
| UniEvalType=Misc2023.05 | 66.7 | |
| BLEUType=Similarity Matching2023.05 | 64.9 | |
| CTCType=Misc2023.05 | 64.5 | |
| SummaC-CONVType=NLI2023.05 | 63.6 | |
| ROUGE-2Type=Similarity Matching2023.05 | 59.2 | |
| ROUGE-1Type=Similarity Matching2023.05 | 58.2 | |
| ROUGE-LType=Similarity Matching2023.05 | 58.2 | |
| SummaC-ZSType=NLI2023.05 | 56.1 | |
| BERTScoreType=Similarity Matching2023.05 | 51.7 | |
| SimCSEType=Similarity Matching2023.05 | 48.6 | |
| BLEURTType=Regression2023.05 | 45.2 | |
| DAEType=NLI2023.05 | 42.7 | |
| FactCCType=Misc2023.05 | 38.6 | |
| QuestEvalType=QA2023.05 | 36.3 | |
| BLANCType=Misc2023.05 | 25.7 | |
| NER-OverlapType=Similarity Matching2023.05 | 0.3 | |
| FEQAType=QA2023.05 | -3.9 | |
| MNLIType=NLI2023.05 | -11 |