Jailbreak Defense Evaluation on ADVB
3.473Strong-Reject Score (SR)vanilla
Evaluation Results
| Method | Links | |
|---|---|---|
| vanillaBase Model=Mistral-7b-U2026.02 | 3.473 | |
| vanillaBase Model=Llama2-7b-U2026.02 | 2.754 | |
| SAGEBase Model=Mistral-7b-U2026.02 | 2.75 | |
| SCANSBase Model=Mistral-7b-U2026.02 | 2.588 | |
| SAGEBase Model=Llama2-7b-U2026.02 | 2.223 | |
| vanillaBase Model=Llama3.1-8b-D2026.02 | 2.067 | |
| ICDBase Model=Llama2-7b-U2026.02 | 1.998 | |
| SAGEBase Model=Llama3.1-8b-D2026.02 | 1.725 | |
| InstructBase Model=Mistral-7b-U2026.02 | 1.658 | |
| SCANSBase Model=Llama2-7b-U2026.02 | 1.604 | |
| SelfDefenDBase Model=Llama3.1-8b-D2026.02 | 1.519 | |
| SelfDefenDBase Model=Llama2-7b-U2026.02 | 1.485 | |
| SCANSBase Model=Llama3.1-8b-D2026.02 | 1.473 | |
| InstructBase Model=Llama3.1-8b-D2026.02 | 1.179 | |
| ICDBase Model=Mistral-7b-U2026.02 | 1.152 | |
| AISABase Model=Llama2-7b-U2026.02 | 1.11 | |
| SelfDefenDBase Model=Mistral-7b-U2026.02 | 1.067 | |
| ICDBase Model=Llama3.1-8b-D2026.02 | 1.029 | |
| AISABase Model=Llama3.1-8b-D2026.02 | 1.027 | |
| AISABase Model=Mistral-7b-U2026.02 | 1.008 | |
| InstructBase Model=Llama2-7b-U2026.02 | 1.004 |