Factual Consistency Evaluation on FRANK-XSum (test)
38.3Pearson Correlation CoefficientALIGNSCORE-large
Evaluation Results
| Method | Links | |
|---|---|---|
| ALIGNSCORE-largeType=Ours2023.05 | 38.3 | |
| ALIGNSCORE-baseType=Ours2023.05 | 33.2 | |
| DAEType=NLI2023.05 | 32.9 | |
| QAFactEvalType=QA2023.05 | 32.1 | |
| UniEvalType=Misc2023.05 | 27.2 | |
| BARTScoreType=Misc2023.05 | 23.7 | |
| CTCType=Misc2023.05 | 20 | |
| MNLIType=NLI2023.05 | 19.7 | |
| QuestEvalType=QA2023.05 | 19.5 | |
| SummaC-CONVType=NLI2023.05 | 17.6 | |
| BLEURTType=Regression2023.05 | 15.6 | |
| SummaC-ZSType=NLI2023.05 | 14.7 | |
| SimCSEType=Similarity Matching2023.05 | 13.5 | |
| BERTScoreType=Similarity Matching2023.05 | 13 | |
| NER-OverlapType=Similarity Matching2023.05 | 11.4 | |
| BLEUType=Similarity Matching2023.05 | 8.7 | |
| FactCCType=Misc2023.05 | 8.3 | |
| ROUGE-LType=Similarity Matching2023.05 | 8 | |
| ROUGE-1Type=Similarity Matching2023.05 | 6.8 | |
| BLANCType=Misc2023.05 | 6.4 | |
| ROUGE-2Type=Similarity Matching2023.05 | 4.9 | |
| FEQAType=QA2023.05 | 3 |