List your company

Safety Evaluation on RMCBench

1.13ASRCodeShield
-2.24620.54243.3366.118Jun 10, 2026

Evaluation results

MethodLinks
CodeShieldBackbone=Qwen2.5-Coder-7B, Attack Configuration=w/o CodeSpear2026.061.130.55
Safe-DPOBackbone=Qwen2.5-Coder-7B, Attack Configuration=w/o CodeSpear2026.063.662.2
CodeShieldBackbone=LLaMA3-8B, Attack Configuration=w/o CodeSpear2026.064.951.83
CodeShieldBackbone=LLaMA3-8B, Attack Configuration=w/ CodeSpear2026.066.043.66
CodeShieldBackbone=Qwen2.5-Coder-7B, Attack Configuration=w/ CodeSpear2026.067.694.4
CodeShieldBackbone=Qwen2.5-7B, Attack Configuration=w/ CodeSpear2026.067.883.11
Safe-DPOBackbone=LLaMA3-8B, Attack Configuration=w/o CodeSpear2026.0611.542.75
CodeShieldBackbone=Qwen2.5-7B, Attack Configuration=w/o CodeSpear2026.0623.6310.26
VanillaBackbone=Qwen2.5-Coder-7B, Attack Configuration=w/o CodeSpear2026.0626.9223.26
Safe-DPOBackbone=Qwen2.5-7B, Attack Configuration=w/o CodeSpear2026.0636.2619.78
Safe-DPOBackbone=LLaMA3-8B, Attack Configuration=w/ CodeSpear2026.0648.934.62
Safe-DPOBackbone=Qwen2.5-7B, Attack Configuration=w/ CodeSpear2026.0652.9333.15
VanillaBackbone=LLaMA3-8B, Attack Configuration=w/o CodeSpear2026.0660.2636.45
VanillaBackbone=LLaMA3-8B, Attack Configuration=w/ CodeSpear2026.0663.3738.83
VanillaBackbone=Qwen2.5-7B, Attack Configuration=w/o CodeSpear2026.0674.1847.62
Safe-DPOBackbone=Qwen2.5-Coder-7B, Attack Configuration=w/ CodeSpear2026.0678.2151.83
VanillaBackbone=Qwen2.5-Coder-7B, Attack Configuration=w/ CodeSpear2026.0682.7862.09
VanillaBackbone=Qwen2.5-7B, Attack Configuration=w/ CodeSpear2026.0685.5363.37