Safety Evaluation on XsTest (harmful rate)
2Harmful RateSafeReAct
Evaluation Results
| Method | Links | |
|---|---|---|
| SafeReActModel=R1-8B2026.03 | 2 | |
| SafeReActModel=R1-14B2026.03 | 2 | |
| SafeReActModel=OT-7B2026.03 | 4 | |
| Circuit-BreakerModel=R1-14B2026.03 | 4 | |
| SafeReActModel=R1-7B2026.03 | 5 | |
| SafeChainModel=R1-14B2026.03 | 5 | |
| SafeChainModel=R1-7B2026.03 | 7 | |
| OriginalModel=R1-8B2026.03 | 8 | |
| OriginalModel=R1-14B2026.03 | 8 | |
| SafeChainModel=R1-8B2026.03 | 9 | |
| SafeChainModel=OT-7B2026.03 | 10 | |
| Circuit-BreakerModel=R1-7B2026.03 | 12 | |
| Circuit-BreakerModel=R1-8B2026.03 | 16 | |
| OriginalModel=OT-7B2026.03 | 19 | |
| Circuit-BreakerModel=OT-7B2026.03 | 20 | |
| OriginalModel=R1-7B2026.03 | 27 |