Factual Consistency Evaluation on Q-X
46.8Kendall's tauALIGNSCORE-large
Evaluation Results
| Method | Links | |
|---|---|---|
| ALIGNSCORE-largeType=Ours2023.05 | 46.8 | |
| ALIGNSCORE-baseType=Ours2023.05 | 42.5 | |
| UniEvalType=Misc2023.05 | 40.9 | |
| DAEType=NLI2023.05 | 37.5 | |
| SummaC-CONVType=NLI2023.05 | 36.8 | |
| QAFactEvalType=QA2023.05 | 36.2 | |
| SummaC-ZSType=NLI2023.05 | 35.8 | |
| NER-OverlapType=Similarity Matching2023.05 | 31.2 | |
| FactCCType=Misc2023.05 | 28.8 | |
| CTCType=Misc2023.05 | 25.1 | |
| ROUGE-1Type=Similarity Matching2023.05 | 14.8 | |
| ROUGE-2Type=Similarity Matching2023.05 | 14.5 | |
| BARTScoreType=Misc2023.05 | 13.9 | |
| ROUGE-LType=Similarity Matching2023.05 | 13.6 | |
| BLEURTType=Regression2023.05 | 10.1 | |
| QuestEvalType=QA2023.05 | 9.7 | |
| SimCSEType=Similarity Matching2023.05 | 9.1 | |
| BLEUType=Similarity Matching2023.05 | 9 | |
| BLANCType=Misc2023.05 | 1.3 | |
| MNLIType=NLI2023.05 | 0.5 | |
| FEQAType=QA2023.05 | -5.3 | |
| BERTScoreType=Similarity Matching2023.05 | -7.3 |