Factual Consistency Evaluation on FRANK CNNDM (test)
67.7PCCALIGNSCORE-large
Evaluation Results
| Method | Links | |
|---|---|---|
| ALIGNSCORE-largeType=Ours2023.05 | 67.7 | |
| QAFactEvalType=QA2023.05 | 64.6 | |
| ALIGNSCORE-baseType=Ours2023.05 | 60 | |
| SummaC-CONVType=NLI2023.05 | 58.7 | |
| UniEvalType=Misc2023.05 | 58.3 | |
| CTCType=Misc2023.05 | 54.5 | |
| SummaC-ZSType=NLI2023.05 | 53.7 | |
| BARTScoreType=Misc2023.05 | 51.9 | |
| BERTScoreType=Similarity Matching2023.05 | 51.7 | |
| QuestEvalType=QA2023.05 | 46.5 | |
| DAEType=NLI2023.05 | 40.5 | |
| ROUGE-2Type=Similarity Matching2023.05 | 38.7 | |
| ROUGE-LType=Similarity Matching2023.05 | 37.7 | |
| BLEURTType=Regression2023.05 | 37.5 | |
| ROUGE-1Type=Similarity Matching2023.05 | 37.1 | |
| BLEUType=Similarity Matching2023.05 | 36.6 | |
| FactCCType=Misc2023.05 | 34.8 | |
| SimCSEType=Similarity Matching2023.05 | 34.5 | |
| BLANCType=Misc2023.05 | 34.3 | |
| NER-OverlapType=Similarity Matching2023.05 | 30.1 | |
| FEQAType=QA2023.05 | -0.4 | |
| MNLIType=NLI2023.05 | -2.2 |