Factual Consistency Evaluation on Q-C
61.3Kendall's TauALIGNSCORE-large
Evaluation Results
| Method | Links | |
|---|---|---|
| ALIGNSCORE-largeType=Ours2023.05 | 61.3 | |
| BARTScoreType=Misc2023.05 | 55.6 | |
| ALIGNSCORE-baseType=Ours2023.05 | 55.4 | |
| UniEvalType=Misc2023.05 | 54.4 | |
| QAFactEvalType=QA2023.05 | 50.5 | |
| SummaC-CONVType=NLI2023.05 | 46.5 | |
| CTCType=Misc2023.05 | 45.7 | |
| BLEUType=Similarity Matching2023.05 | 44.7 | |
| ROUGE-2Type=Similarity Matching2023.05 | 44.2 | |
| ROUGE-1Type=Similarity Matching2023.05 | 42.9 | |
| ROUGE-LType=Similarity Matching2023.05 | 42.8 | |
| SummaC-ZSType=NLI2023.05 | 40.5 | |
| FactCCType=Misc2023.05 | 37.7 | |
| SimCSEType=Similarity Matching2023.05 | 36.7 | |
| BERTScoreType=Similarity Matching2023.05 | 36.3 | |
| DAEType=NLI2023.05 | 34.7 | |
| BLEURTType=Regression2023.05 | 33.9 | |
| QuestEvalType=QA2023.05 | 23.9 | |
| BLANCType=Misc2023.05 | 17.1 | |
| NER-OverlapType=Similarity Matching2023.05 | 0.2 | |
| FEQAType=QA2023.05 | -5.7 | |
| MNLIType=NLI2023.05 | -12.8 |