List your company

Safety Classification on HoliSafe-Bench

0.783AUROCQwen3-VL 8B
0.307720.431110.55450.67789May 1, 2026

Evaluation results

MethodLinks
Qwen3-VL 8BStrategy=Max152026.050.783
Qwen3-VL 4BStrategy=Max152026.050.778
Qwen3-VL 4BStrategy=Sel2026.050.768
Qwen3-VL 4BStrategy=Rand2026.050.764
Qwen3-VL 8BStrategy=Rand2026.050.762
InternVLStrategy=Max152026.050.761
InternVLStrategy=Rand2026.050.758
Qwen3-VL 8BStrategy=Med152026.050.753
Qwen3-VL 4BStrategy=MeanEns2026.050.75
Qwen3-VL 4BStrategy=Med152026.050.748
MiniCPMStrategy=Max152026.050.744
Qwen3-VL 8BStrategy=MeanEns2026.050.739
GemmaStrategy=Max152026.050.737
InternVLStrategy=MeanEns2026.050.731
Qwen3-VL 4BStrategy=Mean152026.050.728
GemmaStrategy=Rand2026.050.727
GemmaStrategy=Med152026.050.725
Qwen3-VL 8BStrategy=Mean152026.050.724
InternVLStrategy=Sel2026.050.72
InternVLStrategy=Med152026.050.72
MiniCPMStrategy=MeanEns2026.050.718
MiniCPMStrategy=Rand2026.050.713
InternVLStrategy=Mean152026.050.701
Qwen30BStrategy=Max152026.050.693
GemmaStrategy=Mean152026.050.689
Qwen30BStrategy=Rand2026.050.688
GemmaStrategy=MeanEns2026.050.688
MiniCPMStrategy=Med152026.050.684
Qwen30BStrategy=MeanEns2026.050.668
Qwen30BStrategy=Med152026.050.667
MiniCPMStrategy=Mean152026.050.663
Qwen30BStrategy=Mean152026.050.657
LlamaStrategy=Max152026.050.65
MiniCPMStrategy=Sel2026.050.648
Qwen30BStrategy=Sel2026.050.638
LlamaStrategy=MeanEns2026.050.603
LlamaStrategy=Rand2026.050.599
LlamaStrategy=Med152026.050.588
Qwen30BStrategy=Min152026.050.584
LlamaStrategy=Mean152026.050.573
Qwen3-VL 4BStrategy=Min152026.050.511
LlamaStrategy=Sel2026.050.504
Qwen3-VL 8BStrategy=Sel2026.050.499
Qwen3-VL 8BStrategy=Min152026.050.499
MiniCPMStrategy=Min152026.050.49
LlamaStrategy=Min152026.050.465
InternVLStrategy=Min152026.050.39
GemmaStrategy=Sel2026.050.326
GemmaStrategy=Min152026.050.326