Multimodal Safety Evaluation on multimodal safety dataset
0.13ASRClaude 3.5 Sonnet
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Claude 3.5 Sonnet2025.04 | 0.13 | 0.6944 | 0.59 | 0.56 | 0.68 | |
| GPT-4o Mini2025.04 | 0.25 | 0.6444 | 0.55 | 0.54 | 0.63 | |
| Gemini 1.5 Pro2025.04 | 0.255 | 0.2333 | 0.76 | 0.76 | 0.75 | |
| Claude 3 Haiku2025.04 | 0.265 | 0.4722 | 0.63 | 0.61 | 0.67 | |
| GPT-4o2025.04 | 0.44 | 0.1278 | 0.72 | 0.81 | 0.66 | |
| Gemini 1.5 Flash2025.04 | 0.635 | 0.2389 | 0.56 | 0.6 | 0.46 |