Factuality Evaluation on Rank19
83.9AccuracyHuman
Evaluation Results
| Method | Links | |
|---|---|---|
| Human2021.06 | 83.9 | |
| BARTSCOREFine-tuning task=CNN2021.06 | 83.6 | |
| BARTSCOREFine-tuning task=CNN, Prompting=true2021.06 | 79.6 | |
| BARTSCOREFine-tuning task=CNN + Para2021.06 | 78.8 | |
| PRISM2021.06 | 78 | |
| QAGS2021.06 | 72.1 | |
| BERTScore2021.06 | 71.3 | |
| MoverScore2021.06 | 71.3 | |
| FactCC2021.06 | 70 | |
| BARTSCORESetup=Vanilla2021.06 | 68.4 | |
| ROUGE-22021.06 | 63 | |
| ROUGE-L2021.06 | 58.7 | |
| ROUGE-12021.06 | 56.8 |