Fact Verification on TrendFact (test)
78.98Accuracyo1-preview
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| o1-previewEvidence Condition=Gold Evidence2024.10 | 78.98 | 75.16 | 75.13 | 75.72 | |
| DeepSeek-R1Evidence Condition=Gold Evidence2024.10 | 77.92 | 72.56 | 73.72 | 72.64 | |
| GPT-4.1Evidence Condition=Gold Evidence2024.10 | 72.29 | 69.68 | 69.02 | 72.88 | |
| QwQ-32B-PreviewEvidence Condition=Gold Evidence2024.10 | 65.31 | 61.76 | 63.68 | 65.53 | |
| Qwen-72B-instructEvidence Condition=Gold Evidence2024.10 | 65.14 | 60.56 | 66.97 | 63.65 | |
| DeepSeek-V3Evidence Condition=Gold Evidence2024.10 | 63.74 | 60.31 | 66.09 | 63.96 | |
| CLAIMDECOMPEvidence Condition=Gold Evidence2024.10 | 59.35 | 56.86 | 56.65 | 59.41 | |
| PROGRAM-FCEvidence Condition=Gold Evidence2024.10 | 56.55 | 54.05 | 54.17 | 56.62 |