Safety evaluation on AdvBench 50 examples
100Safe Response Rate
Direct Instruction
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 100 | |
2025.09 | 98 | |
2025.09 | 98 | |
2025.09 | 96 | |
2025.09 | 96 | |
2025.09 | 94 | |
2025.09 | 92 | |
2025.09 | 90 | |
2025.09 | 88 | |
2025.09 | 86 | |
2025.09 | 70 | |
2025.09 | 60 | |
2025.09 | 58 | |
2025.09 | 56 | |
2025.09 | 48 | |
2025.09 | 48 | |
2025.09 | 40 | |
2025.09 | 36 | |
2025.09 | 18 | |
2025.09 | 2 | |
2025.09 | 2 | |
2025.09 | 0 | |
2025.09 | 0 |