Factuality Evaluation on AggreFact-XSum (EXF)
0.799Balanced AccuracyAlignScore
Evaluation results
| Method | Links | |
|---|---|---|
| AlignScore2024.03 | 0.799 | |
| FENICEGPT_claimsExtractor=LLM-based claims2024.03 | 0.735 | |
| FENICET5_claimsExtractor=Knowledge-distilled claim extractor2024.03 | 0.707 | |
| ChatGPT-StarPrompting=Scale-based2024.03 | 0.706 | |
| ChatGPT-ZSPrompting=Zero-shot2024.03 | 0.692 | |
| TrueTeacher-11BParameters=11B2024.03 | 0.684 | |
| ChatGPT-DAPrompting=Direct Assessment2024.03 | 0.656 | |
| SummaC-Cvvariant=Convolutional2024.03 | 0.646 | |
| ChatGPT-CoTPrompting=Chain-of-Thought2024.03 | 0.609 | |
| QuestEval2024.03 | 0.601 | |
| MENLI2024.03 | 0.597 | |
| QAFactEval2024.03 | 0.596 | |
| SummaC-ZSvariant=Zero-shot2024.03 | 0.514 | |
| Random Baseline2024.03 | 0.5 |