List your company

Malicious Prompt Detection on GPT4V-Caption (IOD)

0FPRVLMShield
-0.31761.82623.976.1138Apr 7, 2026

Evaluation results

MethodLinks
VLMShieldTarget Model=LLaVA-1.5-13B2026.040
VLMShieldTarget Model=Qwen2.5-VL-7B-Instruct2026.040
CIDERTarget Model=LLaVA-1.5-13B2026.042.2
CIDERTarget Model=Qwen2.5-VL-7B-Instruct2026.042.2
ASTRATarget Model=Qwen2.5-VL-7B-Instruct2026.042.26
JailGuardTarget Model=Qwen2.5-VL-7B-Instruct2026.042.64
VLMGuardTarget Model=Qwen2.5-VL-7B-Instruct2026.042.67
ECSOTarget Model=Qwen2.5-VL-7B-Instruct2026.043.7
ASTRATarget Model=LLaVA-1.5-13B2026.043.85
VLMGuardTarget Model=LLaVA-1.5-13B2026.044.76
JailGuardTarget Model=LLaVA-1.5-13B2026.044.91
ECSOTarget Model=LLaVA-1.5-13B2026.046.02
MirrorCheckTarget Model=LLaVA-1.5-13B2026.047.94
MirrorCheckTarget Model=Qwen2.5-VL-7B-Instruct2026.047.94