List your company

Risk Identification on MSSBench

96.05RIRGPT-4o-mini
48.15860.591573.02585.4585Mar 15, 2026

Evaluation results

MethodLinks
GPT-4o-miniModel Category=Proprietary Models2026.0396.0584.21
Gemini-3-ProModel Category=Proprietary Models2026.0385.5310.53
HomeGuard-8BModel Category=Ours, Parameters=8B2026.0377.6325
HomeGuard-4BModel Category=Ours, Parameters=4B2026.0364.4721.05
Qwen3-VL-235B-A22B-ThinkingModel Category=Open-Sourced VLMs2026.0362.6722.37
RoboBrain2.5-8BModel Category=Open-Sourced VLMs2026.0359.2122.37
Gemini-2.5-flashModel Category=Proprietary Models2026.0353.9528.95
PoexModel Category=Runtime Safeguard2026.0353.9527.63
Qwen3-VL-8B-ThinkingModel Category=Open-Sourced VLMs2026.0352.6327.11
Qwen3-VL-4B-ThinkingModel Category=Open-Sourced VLMs2026.0351.3123.68
ThinkSafeModel Category=Runtime Safeguard2026.035030.26
AgentSpecModel Category=Runtime Safeguard2026.035025