List your company

Adversarial Attack on AdvBench (test)

100ASRAutoDAN
2.000827.442952.88578.3271May 9, 2026May 10, 2026May 11, 2026May 12, 2026

Evaluation results

MethodLinks
AutoDANTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=32026.05100———
Semantic Representation AttackTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=32026.05100———
Semantic Representation AttackTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=32026.05100———
Semantic Representation AttackTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=12026.05100———
Semantic Representation AttackTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=32026.05100———
AutoDANTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=32026.05100———
Semantic Representation AttackTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=12026.05100———
Semantic Representation AttackTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=32026.05100———
Semantic Representation AttackTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=52026.05100———
PCAPType=MoE, Attacker Model=Kimi K2.5, Personas=6, Strategies per persona=22026.0510097.54.1888
Semantic Representation AttackTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=12026.0599.81———
AutoDANTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=32026.0599.81———
Semantic Representation AttackTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=52026.0599.81———
Semantic Representation AttackTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=52026.0599.62———
Semantic Representation AttackTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=52026.0599.62———
Semantic Representation AttackTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=52026.0599.62———
Semantic Representation AttackTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=32026.0599.62———
Semantic Representation AttackTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=52026.0599.62———
AutoDANTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=52026.0599.42———
Semantic Representation AttackTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=32026.0599.42———
AutoDANTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=52026.0599.42———
AutoDANTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=52026.0599.42———
PCAPType=MoE, Attacker Model=Mixtral 8x22B, Personas=6, Strategies per persona=22026.0599.396.32.4999.09
PCAPType=Dense, Attacker Model=Granite 3.3 8B, Personas=6, Strategies per persona=22026.0598.795.31.98169.06
PCAPType=MoE, Attacker Model=Granite 4.0 H Tiny, Personas=6, Strategies per persona=22026.059894.32.41135.6
Semantic Representation AttackTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=52026.0597.31———
PCAPType=Dense, Attacker Model=Llama 3.3 70B, Personas=6, Strategies per persona=22026.0597.393.32.9890.26
Semantic Representation AttackTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=32026.0596.92———
Semantic Representation AttackTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=32026.0596.73———
Semantic Representation AttackTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=52026.0596.73———
Semantic Representation AttackTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=52026.0595.96———
Semantic Representation AttackTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=32026.0595.19———
TAPType=MoE, Attacker Model=Kimi K2.52026.0594891.2429.8
Semantic Representation AttackTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=12026.0591.73———
Semantic Representation AttackTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=12026.0590.58———
Semantic Representation AttackTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=32026.0590———
Semantic Representation AttackTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=52026.0589.62———
TAPType=Dense, Attacker Model=Granite 3.3 8B2026.0588.782.61.0537.16
Semantic Representation AttackTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=12026.0588.27———
Semantic Representation AttackTime Budget=30s, Target Model=Vicuna 13B, Defense Intensity=52026.0587.69———
Semantic Representation AttackTime Budget=30s, Target Model=Vicuna 13B, Defense Intensity=32026.0587.12———
Semantic Representation AttackTime Budget=15s, Target Model=Vicuna 13B, Defense Intensity=52026.0585.19———
Semantic Representation AttackTime Budget=15s, Target Model=Vicuna 13B, Defense Intensity=32026.0584.04———
BEASTTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=52026.0583.85———
Semantic Representation AttackTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=12026.0579.81———
AutoDANTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=52026.0578.65———
BEASTTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=52026.0578.65———
AutoDANTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=52026.0578.27———
AutoDANTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=52026.0578.27———
AutoDANTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=52026.0578.27———
AutoDANTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=52026.0577.5———
AutoDANTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=52026.0577.12———
Semantic Representation AttackTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=12026.0576.54———
AutoDANTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=32026.0574.81———
AutoDANTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=32026.0570.38———
AutoDANTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=32026.0569.81———
TAPType=MoE, Attacker Model=Mixtral 8x22B2026.0569.361.50.7953.13
BEASTTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=52026.0567.31———
BEASTTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=52026.0566.73———
BEASTTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=32026.0564.04———
BEASTTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=52026.0563.85———
TAPType=Dense, Attacker Model=Llama 3.3 70B2026.0557.748.70.7256.24
BEASTTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=32026.0550.96———
BEASTTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=32026.0547.5———
Semantic Representation AttackTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=12026.0544.62———
BEASTTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=52026.0544.04———
AutoDANTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=32026.0542.69———
TAPType=MoE, Attacker Model=Granite 4.0 H Tiny2026.054133.80.4369.49
AutoDANTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=32026.0538.85———
AutoDANTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=32026.0538.46———
AutoDANTime Budget=30s, Target Model=Vicuna 13B, Defense Intensity=52026.0538.27———
BEASTTime Budget=30s, Target Model=Vicuna 7B, Defense Intensity=32026.0534.62———
BEASTTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=32026.0534.62———
AutoDANTime Budget=15s, Target Model=Vicuna 13B, Defense Intensity=52026.0534.42———
BEASTTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=52026.0534.23———
BEASTTime Budget=30s, Target Model=Vicuna 13B, Defense Intensity=52026.0532.88———
AutoDANTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=52026.0531.73———
GCGTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=52026.0531.54———
BEASTTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=52026.0530.96———
BEASTTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=52026.0529.04———
BEASTTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=52026.0526.54———
BEASTTime Budget=15s, Target Model=Mistral 7B, Defense Intensity=32026.0525.19———
BEASTTime Budget=15s, Target Model=Vicuna 13B, Defense Intensity=52026.0523.85———
BEASTTime Budget=30s, Target Model=Mistral 7B, Defense Intensity=32026.0521.15———
BEASTTime Budget=60s, Target Model=Vicuna 7B, Defense Intensity=32026.0515.77———
BEASTTime Budget=15s, Target Model=Vicuna 13B, Defense Intensity=32026.0515.19———
BEASTTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=32026.0514.04———
BEASTTime Budget=30s, Target Model=Vicuna 13B, Defense Intensity=32026.0513.46———
GCGTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=32026.059.23———
AutoDANTime Budget=15s, Target Model=Vicuna 13B, Defense Intensity=32026.059.23———
BEASTTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=12026.058.65———
AutoDANTime Budget=30s, Target Model=Vicuna 13B, Defense Intensity=32026.058.27———
AutoDANTime Budget=15s, Target Model=Guanaco 7B, Defense Intensity=12026.058.08———
AutoDANTime Budget=30s, Target Model=Guanaco 7B, Defense Intensity=12026.058.08———
AutoDANTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=32026.057.88———
AutoDANTime Budget=60s, Target Model=Guanaco 7B, Defense Intensity=12026.057.88———
BEASTTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=32026.057.69———
GCGTime Budget=60s, Target Model=Mistral 7B, Defense Intensity=52026.056.54———
Semantic Representation AttackTime Budget=60s, Target Model=Vicuna 13B, Defense Intensity=12026.056.15———
BEASTTime Budget=15s, Target Model=Vicuna 7B, Defense Intensity=12026.055.77———