List your company

Safety Classification on UnsafeBench

80.5AUROCGemma
33.28445.54257.870.058May 1, 2026

Evaluation results

MethodLinks
GemmaStrategy=Max152026.0580.5
Qwen3-VL 8BStrategy=Max152026.0580
InternVLStrategy=Max152026.0579.4
GemmaStrategy=Med152026.0579.2
InternVLStrategy=MeanEns2026.0579
Qwen30BStrategy=Max152026.0578.9
Qwen3-VL 4BStrategy=Max152026.0578.4
MiniCPMStrategy=Max152026.0578.4
InternVLStrategy=Med152026.0578.1
GemmaStrategy=MeanEns2026.0578
Qwen3-VL 8BStrategy=Med152026.0577.4
Qwen30BStrategy=Sel2026.0577.2
GemmaStrategy=Rand2026.0576.9
Qwen3-VL 8BStrategy=MeanEns2026.0576.6
Qwen30BStrategy=Med152026.0576.6
Qwen30BStrategy=MeanEns2026.0576.3
Qwen3-VL 4BStrategy=MeanEns2026.0576.2
InternVLStrategy=Rand2026.0576.2
MiniCPMStrategy=MeanEns2026.0576.1
GemmaStrategy=Mean152026.0575.6
Qwen3-VL 4BStrategy=Rand2026.0575.5
Qwen3-VL 4BStrategy=Med152026.0575.4
Qwen3-VL 8BStrategy=Mean152026.0575.1
InternVLStrategy=Mean152026.0575.1
Qwen3-VL 4BStrategy=Sel2026.0574.8
Qwen3-VL 8BStrategy=Rand2026.0574.7
Qwen3-VL 4BStrategy=Mean152026.0573.6
Qwen30BStrategy=Rand2026.0573.6
Qwen30BStrategy=Mean152026.0573.4
InternVLStrategy=Sel2026.0572.8
LlamaStrategy=Max152026.0571.2
LlamaStrategy=MeanEns2026.0570.8
MiniCPMStrategy=Rand2026.0569.8
MiniCPMStrategy=Med152026.0569.8
MiniCPMStrategy=Mean152026.0569.1
MiniCPMStrategy=Sel2026.0568.5
LlamaStrategy=Sel2026.0565.2
LlamaStrategy=Med152026.0565.2
LlamaStrategy=Mean152026.0563.4
LlamaStrategy=Rand2026.0559.5
LlamaStrategy=Min152026.0551.5
MiniCPMStrategy=Min152026.0549.4
Qwen30BStrategy=Min152026.0549.3
Qwen3-VL 4BStrategy=Min152026.0548.6
InternVLStrategy=Min152026.0546.7
Qwen3-VL 8BStrategy=Sel2026.0543
Qwen3-VL 8BStrategy=Min152026.0543
GemmaStrategy=Sel2026.0535.1
GemmaStrategy=Min152026.0535.1