Safety Control on DeepSeek-R1-Distill-Qwen-1.5B
89.8P_safeguarded (Safety-Quality Score)
SafeCtrl-RL
Updated 8d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.05 | 89.8 | 0.608 | |
2026.05 | 87.9 | 0.589 | |
2026.05 | 84.3 | 0.552 | |
2026.05 | 83.2 | 0.542 | |
2026.05 | 82.8 | 0.538 | |
2026.05 | 78.1 | 0.49 | |
2026.05 | 77.7 | 0.486 | |
2026.05 | 75.3 | 0.463 | |
2026.05 | 73.8 | 0.448 | |
2026.05 | 64.5 | 0.355 | |
2026.05 | 61.5 | 0.325 | |
2026.05 | 54.9 | 0.259 | |
2026.05 | 16.7 | -0.123 | |
2026.05 | 15.6 | -0.134 | |
2026.05 | 10 | -0.19 | |
2026.05 | 6.9 | -0.221 | |
2026.05 | 6.5 | -0.225 |