Keyword Prediction on AutoDAN jailbreaking prompts (test)
43.4Precision
EnsembleSHAP
Updated 3mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 43.4 | 37.9 | |
2026.03 | 33.5 | 28.1 | |
2026.03 | 30.6 | 26 | |
2026.03 | 25.2 | 21.2 |
EnsembleSHAP
| Method | Links | ||
|---|---|---|---|
2026.03 | 43.4 | 37.9 | |
2026.03 | 33.5 | 28.1 | |
2026.03 | 30.6 | 26 | |
2026.03 | 25.2 | 21.2 |