Long-form Fact-checking on LongForm Bio
0.8568FactScoreSERC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SERCModel=8B2026.04 | 0.8568 | 18.53 | 77.7 | |
| SERCModel=14B2026.04 | 0.8146 | 22.53 | 106.5 | |
| RARRModel=8B2026.04 | 0.7999 | 22.96 | 96.2 | |
| RARRModel=14B2026.04 | 0.7647 | 21.21 | 100.2 | |
| Re-ExModel=8B2026.04 | 0.7636 | 23.53 | 98.6 | |
| CoVe+RAGModel=14B2026.04 | 0.7427 | 17.54 | 82.9 | |
| CoVe+RAGModel=8B2026.04 | 0.7265 | 17.7 | 74.2 | |
| Re-ExModel=14B2026.04 | 0.7239 | 21.77 | 102.8 | |
| Initial AnswerModel=8B2026.04 | 0.5976 | 23.86 | 100 | |
| CoVeModel=8B2026.04 | 0.5551 | 17.23 | 72.2 | |
| Initial AnswerModel=14B2026.04 | 0.485 | 21.16 | 100 | |
| CoVeModel=14B2026.04 | 0.4625 | 14.85 | 70.2 |