Factuality Evaluation on AggreFact-XSum FTS
80.2Balanced AccuracyAlignScore
Evaluation Results
| Method | Links | |
|---|---|---|
| AlignScore2024.03 | 80.2 | |
| TrueTeacher-11BParameters=11B2024.03 | 75.2 | |
| FENICEGPT_claimsExtractor=LLM-based claims2024.03 | 73.9 | |
| DAENote=Trained on XSumFaith2024.03 | 73.1 | |
| FENICET5_claimsExtractor=Knowledge-distilled claim extractor2024.03 | 71.4 | |
| SummaC-Cvvariant=Convolutional2024.03 | 67 | |
| QAFactEval2024.03 | 65.9 | |
| ChatGPT-ZSPrompting=Zero-shot2024.03 | 62.6 | |
| QuestEval2024.03 | 61.6 | |
| MENLI2024.03 | 58.3 | |
| ChatGPT-StarPrompting=Scale-based2024.03 | 57.7 | |
| SummaC-ZSvariant=Zero-shot2024.03 | 56.1 | |
| ChatGPT-CoTPrompting=Chain-of-Thought2024.03 | 56 | |
| ChatGPT-DAPrompting=Direct Assessment2024.03 | 53.6 | |
| Random Baseline2024.03 | 50 |