Factuality Evaluation on AggreFact CNN (EXF)
76.5Balanced AccuracySummaC-ZS
Evaluation Results
| Method | Links | |
|---|---|---|
| SummaC-ZSvariant=Zero-shot2024.03 | 76.5 | |
| AlignScore2024.03 | 73.9 | |
| SummaC-Cvvariant=Convolutional2024.03 | 69.8 | |
| QAFactEval2024.03 | 69.1 | |
| FENICEGPT_claimsExtractor=LLM-based claims2024.03 | 68.8 | |
| DAENote=Trained on XSumFaith2024.03 | 67.9 | |
| FENICET5_claimsExtractor=Knowledge-distilled claim extractor2024.03 | 67.8 | |
| ChatGPT-StarPrompting=Scale-based2024.03 | 65.8 | |
| ChatGPT-ZSPrompting=Zero-shot2024.03 | 64.5 | |
| QuestEval2024.03 | 64.3 | |
| ChatGPT-DAPrompting=Direct Assessment2024.03 | 63.6 | |
| ChatGPT-CoTPrompting=Chain-of-Thought2024.03 | 60.4 | |
| TrueTeacher-11BParameters=11B2024.03 | 57.7 | |
| MENLI2024.03 | 52.8 | |
| Random Baseline2024.03 | 50 |