Jailbreak Defense Evaluation on EMRA HQ
0ASRSecurityLingua
Evaluation Results
| Method | Links | |
|---|---|---|
| SecurityLinguaTarget LLM=GPT-52026.04 | 0 | |
| CoopGuardTarget LLM=GPT-52026.04 | 0 | |
| GoalPriorityTarget LLM=Gemini-2.5-Pro2026.04 | 0 | |
| CoopGuardTarget LLM=Gemini-2.5-Pro2026.04 | 0 | |
| Self-ReminderTarget LLM=DeepSeek-V32026.04 | 0 | |
| CoopGuardTarget LLM=DeepSeek-V32026.04 | 0 | |
| RPOTarget LLM=GPT-52026.04 | 0.3 | |
| Self-ReminderTarget LLM=Gemini-2.5-Pro2026.04 | 0.3 | |
| GoalPriorityTarget LLM=DeepSeek-V32026.04 | 0.6 | |
| GoalPriorityTarget LLM=GPT-52026.04 | 1 | |
| Self-ReminderTarget LLM=GPT-52026.04 | 1.1 | |
| PATTarget LLM=GPT-52026.04 | 1.3 | |
| PATTarget LLM=Gemini-2.5-Pro2026.04 | 2.1 | |
| RPOTarget LLM=Gemini-2.5-Pro2026.04 | 2.1 | |
| PATTarget LLM=DeepSeek-V32026.04 | 3.1 | |
| RPOTarget LLM=DeepSeek-V32026.04 | 4.3 | |
| SecurityLinguaTarget LLM=Gemini-2.5-Pro2026.04 | 9.3 | |
| SecurityLinguaTarget LLM=DeepSeek-V32026.04 | 9.3 |