List your company

Multimodal Jailbreaking on VE-Safety

65ASRMM-Plan
-1.97615.41232.850.188Feb 5, 2026

Evaluation results

MethodLinks
MM-PlanTarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0265
MM-PlanTarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0264.4
MM-PlanTarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0254.4
MM-PlanTarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.0246.3
CrescendoTarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0245
FigStepTarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0244.4
MM-PlanTarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.0243.8
SSATarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0239.1
MM-PlanTarget Model=GPT-4o, Model Access Type=Proprietary2026.0236.9
Direct PlanTarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0234.7
CrescendoTarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0233.8
FigStepTarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0233.1
SI-AttackTarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0231.9
SSATarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0229.4
SI-AttackTarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0229.1
Direct RequestTarget Model=InternVL3-8B, Model Access Type=Open-Weight2026.0227.2
MM-PlanTarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.0227.2
SI-AttackTarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0225.6
SSATarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0225.3
FigStepTarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.0224.4
FigStepTarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0223.8
Direct PlanTarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0222.5
CrescendoTarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0221.9
Direct PlanTarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0218.1
CrescendoTarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.0218.1
SSATarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.0215.9
CrescendoTarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.0215.9
SI-AttackTarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.0215.6
CrescendoTarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.0215
CrescendoTarget Model=GPT-4o, Model Access Type=Proprietary2026.0214.4
MM-PlanTarget Model=GPT-5, Model Access Type=Proprietary2026.0213.8
Direct RequestTarget Model=Llama-3.2-11B, Model Access Type=Open-Weight2026.0213.4
FigStepTarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.0213.4
SI-AttackTarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.0212.8
SI-AttackTarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.0212.5
SSATarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.0212.2
Direct RequestTarget Model=Qwen3-VL-8B, Model Access Type=Open-Weight2026.0211.9
Direct PlanTarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.0211.9
FigStepTarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.0211.3
Direct RequestTarget Model=Gemini 2.5 Pro, Model Access Type=Proprietary2026.029.7
Direct PlanTarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.029.7
SSATarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.029.7
Direct PlanTarget Model=GPT-4o, Model Access Type=Proprietary2026.029.4
Direct RequestTarget Model=Sonnet 4.5, Model Access Type=Proprietary2026.028.4
Direct PlanTarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.028.1
SI-AttackTarget Model=GPT-4o, Model Access Type=Proprietary2026.028.1
FigStepTarget Model=GPT-4o, Model Access Type=Proprietary2026.026.6
SSATarget Model=GPT-4o, Model Access Type=Proprietary2026.026.3
Direct RequestTarget Model=GPT-4o, Model Access Type=Proprietary2026.025
Direct RequestTarget Model=Sonnet 3.7, Model Access Type=Proprietary2026.024.7
CrescendoTarget Model=GPT-5, Model Access Type=Proprietary2026.023.1
SI-AttackTarget Model=GPT-5, Model Access Type=Proprietary2026.021.9
SSATarget Model=GPT-5, Model Access Type=Proprietary2026.021.6
Direct PlanTarget Model=GPT-5, Model Access Type=Proprietary2026.020.9
Direct RequestTarget Model=GPT-5, Model Access Type=Proprietary2026.020.6
FigStepTarget Model=GPT-5, Model Access Type=Proprietary2026.020.6