Prefill-stage hallucination risk detection on Benchmark-500 Strict Consensus Pvote = 1.0 vs. Clean
0.6939AUROC (Mean)
Risk-Cos
Updated 27d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 0.6939 | 0.58 | |
2026.03 | 0.685 | 0.57 | |
2026.03 | 0.5323 | 0.44 | |
2026.03 | 0.3508 | 0.25 |