Hallucination Detection on HaluEval (Macro F1)
42.4Macro F1MiniCheck-FT5
Evaluation Results
| Method | Links | |
|---|---|---|
| MiniCheck-FT5Output format=binary2026.06 | 42.4 | |
| SEVA-SFT (3B)Output format=struct, Size=3B2026.06 | 42 | |
| SEVA-GRPO (3B)Output format=struct, Size=3B2026.06 | 39.4 | |
| GPT-4o-miniOutput format=struct2026.06 | 34 |