Attack Success Rate Evaluation on HRL/LRL Safety Prompts English Text v1
1ASRClaude 3 Haiku
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude 3 Haiku2025.04 | 1 | |
| Claude 3.5 Sonnet2025.04 | 1 | |
| Gemini 1.5 Pro2025.04 | 2 | |
| Gemini 1.5 Flash2025.04 | 3 | |
| GPT-4o Mini2025.04 | 9 | |
| GPT-4o2025.04 | 10 |