Jailbreak Success Rate on med-safety
7.6Success Rate (Raw)GPT-4o
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-4oModel=GPT-4o2025.11 | 7.6 | 2 | 0.2 | 0 | |
| DeepSeek-R1Model=DeepSeek-R12025.11 | 2 | 0.4 | 0 | 0 | |
| Qwen3-235BModel=Qwen3-235B2025.11 | 1.1 | 0 | 0 | 0 |
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-4oModel=GPT-4o2025.11 | 7.6 | 2 | 0.2 | 0 | |
| DeepSeek-R1Model=DeepSeek-R12025.11 | 2 | 0.4 | 0 | 0 | |
| Qwen3-235BModel=Qwen3-235B2025.11 | 1.1 | 0 | 0 | 0 |