Jailbreak Attack on HADES All categories
18.93ASRHADES
Evaluation Results
| Method | Links | |
|---|---|---|
| HADESModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 18.93 | |
| HADESModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 22 | |
| HADESModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 33.87 | |
| HADESModel=Grok-2-vision, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 45.6 | |
| SI-AModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 48 | |
| SI-AModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 49.6 | |
| SI-AModel=Grok-2-vision, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 52.4 | |
| SI-AModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 53.6 | |
| VisCRAModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 56.6 | |
| VisCRAModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 61.2 | |
| VisCRAModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 79.73 | |
| GAMBITModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 85.87 | |
| GAMBITModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 88.4 | |
| GAMBITModel=Grok-2-vision, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 91.07 | |
| GAMBITModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.01 | 91.33 |