List your company

Red Teaming Attack on HarmBench (test)

31.18ZSLlama3.2-3B-Instruct
2.871210.220617.5724.9194Apr 2, 2026

Evaluation results

MethodLinks
Llama3.2-3B-InstructBackbone=Llama3.2-3B-Instruct, Alignment Technique=Instruct2026.0431.1821.6520.3819.6738.4247.17
Qwen2.5-7B-InstructBackbone=Qwen2.5-7B-Instruct, Alignment Technique=Instruct2026.0427.9926.0825.127.463.0467.69
Llama3.1-8B-InstructBackbone=Llama3.1-8B-Instruct, Alignment Technique=Instruct2026.0426.8224.7222.5324.8441.7649.93
PROBackbone=Llama3.2-3B-Instruct, Alignment Technique=PRO2026.0416.58.388.399.6833.7739.46
SFTBackbone=Llama3.2-3B-Instruct, Alignment Technique=SFT2026.0415.389.7410.1111.4235.3542.26
PROBackbone=Llama3.1-8B-Instruct, Alignment Technique=PRO2026.0414.9310.729.6711.5932.4538.74
SFTBackbone=Llama3.1-8B-Instruct, Alignment Technique=SFT2026.0414.3611.4410.3713.9734.6940.31
PRO w/ PLOTBackbone=Llama3.2-3B-Instruct, Alignment Technique=PRO, Defense Configuration=PLOT2026.0413.46.966.37.1127.6236.28
PROBackbone=Qwen2.5-7B-Instruct, Alignment Technique=PRO2026.0413.2611.9412.5114.7437.9751.31
SFTBackbone=Qwen2.5-7B-Instruct, Alignment Technique=SFT2026.0412.9313.3813.8316.1739.9953.44
SFT w/ PLOTBackbone=Llama3.2-3B-Instruct, Alignment Technique=SFT, Defense Configuration=PLOT2026.0411.967.68.949.8730.4740.22
DPOBackbone=Llama3.1-8B-Instruct, Alignment Technique=DPO2026.0410.746.476.157.8927.6233.54
PRO w/ PLOTBackbone=Llama3.1-8B-Instruct, Alignment Technique=PRO, Defense Configuration=PLOT2026.0410.558.258.39.2826.4935.12
SFT w/ PLOTBackbone=Llama3.1-8B-Instruct, Alignment Technique=SFT, Defense Configuration=PLOT2026.049.619.198.6810.9829.5437.12
AOTBackbone=Llama3.1-8B-Instruct, Alignment Technique=AOT2026.048.945.584.886.2424.8629.3
DPOBackbone=Qwen2.5-7B-Instruct, Alignment Technique=DPO2026.048.557.157.8210.6733.6645.01
DPOBackbone=Llama3.2-3B-Instruct, Alignment Technique=DPO2026.048.465.455.856.7525.7530.08
PRO w/ PLOTBackbone=Qwen2.5-7B-Instruct, Alignment Technique=PRO, Defense Configuration=PLOT2026.048.198.478.2211.0828.5846.67
SFT w/ PLOTBackbone=Qwen2.5-7B-Instruct, Alignment Technique=SFT, Defense Configuration=PLOT2026.047.4310.5211.2813.4334.7149.15
AOTBackbone=Qwen2.5-7B-Instruct, Alignment Technique=AOT2026.047.035.994.077.8826.8638.59
DPO w/ PLOTBackbone=Llama3.1-8B-Instruct, Alignment Technique=DPO, Defense Configuration=PLOT2026.046.155.144.215.2620.7828.47
AOT w/ PLOTBackbone=Llama3.1-8B-Instruct, Alignment Technique=AOT, Defense Configuration=PLOT2026.045.974.193.044.5317.9526.65
AOTBackbone=Llama3.2-3B-Instruct, Alignment Technique=AOT2026.045.85.165.656.0321.1128.93
AOT w/ PLOTBackbone=Qwen2.5-7B-Instruct, Alignment Technique=AOT, Defense Configuration=PLOT2026.045.274.893.255.720.2135.27
DPO w/ PLOTBackbone=Qwen2.5-7B-Instruct, Alignment Technique=DPO, Defense Configuration=PLOT2026.044.645.065.537.2725.1339.7
DPO w/ PLOTBackbone=Llama3.2-3B-Instruct, Alignment Technique=DPO, Defense Configuration=PLOT2026.044.394.234.324.9216.9226.83
AOT w/ PLOTBackbone=Llama3.2-3B-Instruct, Alignment Technique=AOT, Defense Configuration=PLOT2026.043.964.434.364.8418.5725.54