Factual Consistency Evaluation on SummEval (test)
66.3Pearson CCALIGNSCORE-large
Evaluation results
| Method | Links | |
|---|---|---|
| ALIGNSCORE-largeType=Ours2023.05 | 66.3 | |
| QAFactEvalType=QA2023.05 | 61.6 | |
| ALIGNSCORE-baseType=Ours2023.05 | 61.1 | |
| UniEvalType=Misc2023.05 | 57.8 | |
| CTCType=Misc2023.05 | 54.7 | |
| SummaC-CONVType=NLI2023.05 | 50.3 | |
| SummaC-ZSType=NLI2023.05 | 50.1 | |
| ROUGE-LType=Similarity Matching2023.05 | 42.3 | |
| BLEUType=Similarity Matching2023.05 | 41.5 | |
| DAEType=NLI2023.05 | 41.5 | |
| ROUGE-1Type=Similarity Matching2023.05 | 41.1 | |
| ROUGE-2Type=Similarity Matching2023.05 | 40.9 | |
| BARTScoreType=Misc2023.05 | 35.5 | |
| FactCCType=Misc2023.05 | 34.8 | |
| BERTScoreType=Similarity Matching2023.05 | 33.1 | |
| QuestEvalType=QA2023.05 | 29.7 | |
| SimCSEType=Similarity Matching2023.05 | 28.5 | |
| NER-OverlapType=Similarity Matching2023.05 | 24.9 | |
| BLEURTType=Regression2023.05 | 23.8 | |
| BLANCType=Misc2023.05 | 21.3 | |
| MNLIType=NLI2023.05 | -1.8 | |
| FEQAType=QA2023.05 | -2.9 |