Fact Verification on FEVER (Macro F1)
91Macro F1GPT-4o-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o-miniOutput format=struct2026.06 | 91 | |
| MiniCheck-FT5Output format=binary2026.06 | 87.1 | |
| SEVA-GRPO (3B)Output format=struct, Size=3B2026.06 | 84.9 | |
| SEVA-SFT (3B)Output format=struct, Size=3B2026.06 | 76.3 |