Jailbreak Detection on JBB-Behaviors Harmful
99.94AccuracySentra-Guard
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Sentra-GuardPrompt Types=100 adversarial behaviors.2025.10 | 99.94 | 100 | 99.89 | 99.94 | 0.7 |
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Sentra-GuardPrompt Types=100 adversarial behaviors.2025.10 | 99.94 | 100 | 99.89 | 99.94 | 0.7 |