List your company

Jailbreak Attack on HADES Self-harm

5.33ASRHADES
1.756425.87825074.1218Jan 6, 2026

Evaluation results

MethodLinks
HADESModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.015.33
HADESModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0113.33
HADESModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0120
SI-AModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0132.67
SI-AModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0132.67
SI-AModel=Grok-2-vision, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0133.33
SI-AModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0135.33
HADESModel=Grok-2-vision, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0138
VisCRAModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0144.67
VisCRAModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0153.33
VisCRAModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0168.67
GAMBITModel=GPT-4o, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0188
GAMBITModel=InternVL 2.5, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0190.67
GAMBITModel=Grok-2-vision, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0192.67
GAMBITModel=Qwen2.5-VL, Evaluation Protocol=Pass@5, Evaluator=Llama-Guard-32026.0194.67