List your company

Jailbreaking on StrongReject (test)

89.8Attack Success Rate (ASR)PAD
-3.59220.65444.969.146Jun 1, 2026

Evaluation results

MethodLinks
PADVictim=Dream-Instruct2026.0689.8———4.47
PADVictim=MMaDA-CoT2026.0687.9———4.5
DIJAVictim=MMaDA-CoT2026.0687.5———4.55
DIJAVictim=Dream-Instruct2026.0687.2———4.64
MASKFORGEVictim=LLaDA-Instruct2026.0687———4.68
AutoDANVictim=MMaDA-CoT2026.0683.5———3.66
DIJAVictim=LLaDA-Instruct2026.0683.1———4.55
MASKFORGEVictim=LLaDA 1.52026.0683———4.63
PADVictim=LLaDA 1.52026.0682.1———4.29
PADVictim=LLaDA-Instruct2026.0681.8———4.42
MASKFORGEVictim=Dream-Instruct2026.0681———4.55
MASKFORGEVictim=AVG2026.0681———4.58
DIJAVictim=LLaDA 1.52026.0680.8———4.51
MASKFORGEVictim=LLaDA2.0-mini2026.0679———4.64
DIRECTVictim=MMaDA-CoT2026.0675.7———4
MASKFORGEVictim=MMaDA-CoT2026.0675———4.4
AutoDANVictim=LLaDA-Instruct2026.0674.5———3.41
DIJAVictim=AVG2026.0674.3———4.34
PAIRVictim=MMaDA-CoT2026.0672.5———3.75
PADVictim=AVG2026.0672.5———4.15
AutoDANVictim=LLaDA 1.52026.0672———3.36
AutoDANVictim=AVG2026.0662.5———3.18
AutoDANVictim=Dream-Instruct2026.0646.5———2.93
AutoDANVictim=LLaDA2.0-mini2026.0636———2.54
DIJAVictim=LLaDA2.0-mini2026.0632.9———3.46
PAIRVictim=LLaDA 1.52026.0632.5———2.04
PAIRVictim=LLaDA-Instruct2026.0631———1.97
PAIRVictim=AVG2026.0630.7———2.11
AIMVictim=MMaDA-CoT2026.0626———2.1
PADVictim=LLaDA2.0-mini2026.0620.8———3.09
DIRECTVictim=AVG2026.0615.7———1.63
AIMVictim=LLaDA-Instruct2026.0612———1.72
AIMVictim=AVG2026.0610.8———1.57
AIMVictim=LLaDA 1.52026.0610.5———1.68
PAIRVictim=LLaDA2.0-mini2026.069.5———1.3
PAIRVictim=Dream-Instruct2026.068———1.47
AIMVictim=LLaDA2.0-mini2026.064———1.2
DIRECTVictim=LLaDA 1.52026.061.6———1.08
AIMVictim=Dream-Instruct2026.061.5———1.15
DIRECTVictim=LLaDA-Instruct2026.060.6———1.04
DIRECTVictim=LLaDA2.0-mini2026.060.3———1.02
DIRECTVictim=Dream-Instruct2026.060———1.02
ADCTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—61.556.410.9—
ADCTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—5550.111.6—
ADCTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—63.958.810.6—
ArrAttackTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—8889.098—
ArrAttackTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—80748.6—
ArrAttackTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—90867.9—
AutoDANTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—66.260.413.6—
AutoDANTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—58.853.214.2—
AutoDANTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—67.962.313.1—
Foot-In-The-Door (FITD)Target LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—38.733.417.1—
Foot-In-The-Door (FITD)Target LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—34.730.117.9—
Foot-In-The-Door (FITD)Target LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—40.135.116.5—
HMNSTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—94892.2—
HMNSTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—86802.2—
HMNSTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—96922.1—
Many-shot JB (MSJ)Target LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—58.353.111.8—
Many-shot JB (MSJ)Target LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—52.647.912.7—
Many-shot JB (MSJ)Target LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—60.855.911.5—
MasterKeyTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—63.658.211.2—
MasterKeyTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—5650.812—
MasterKeyTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—6559.811.1—
PrisonBreakTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—7165.412.4—
PrisonBreakTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—66.96113—
PrisonBreakTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—7266.912.2—
TempestTarget LLM=LLaMA-2-7B-Chat, Zero-shot=true2026.04—78.673.29.9—
TempestTarget LLM=Phi-3-Medium-14B (Instruct), Zero-shot=true2026.04—7670.410.2—
TempestTarget LLM=LLaMA-3.1-70B, Zero-shot=true2026.04—79.274.69.8—