Fact Checking on FEVER
93.1Accuracyhybrid fact-verification pipeline
Evaluation results
| Method | Links | |
|---|---|---|
| hybrid fact-verification pipelineMode=S/R2025.11 | 93.1 | |
| Lewis et al.Mode=S/R2025.11 | 89.5 | |
| MindRefbackbone=LLaMA-2-13b2024.02 | 78.79 | |
| MindRefbackbone=LLaMA-13b2024.02 | 78.53 | |
| MindRefbackbone=LLaMA-7b2024.02 | 78.1 | |
| MindRefbackbone=LLaMA-2-7b2024.02 | 78.04 | |
| BM252024.02 | 77.54 | |
| DPRtraining=full data2024.02 | 75.27 | |
| Contriever2024.02 | 73.61 | |
| LLaMA-2-13b2024.02 | 73.23 | |
| hybrid fact-verification pipelineMode=S/R/N2025.11 | 70.2 | |
| Tan et al.Mode=S/R/N2025.11 | 54.2 |