Safety Guardrail on Prompt-side safety benchmarks
88.6F1 Score
Stage3-DPO
Updated 17d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.07 | 88.6 | |
2026.07 | 86.5 | |
2026.07 | 86.3 | |
2026.07 | 85.9 | |
2026.07 | 85.5 | |
2026.07 | 85.2 | |
2026.07 | 84.9 | |
2026.07 | 84.4 | |
2026.07 | 84 | |
2026.07 | 83.3 | |
2026.07 | 83.1 | |
2026.07 | 82.5 |