List your company

Jailbreak Robustness on STRONGREJECT (train)

93EVUSIHO
-3.7221.3946.571.61Jun 2, 2026

Evaluation results

MethodLinks
IHOTarget model=Qwen-2.5-7B2026.0693
IHOTarget model=Qwen-2.5-32B2026.0691
PAIR+Target model=Qwen-2.5-7B2026.0691
JR1+Target model=Qwen-2.5-7B2026.0688
PAIR+Target model=Qwen-2.5-32B2026.0687
JR1+Target model=Qwen-2.5-32B2026.0687
GCG+Target model=Qwen-2.5-7B2026.0687
GCG+Target model=Qwen-2.5-32B2026.0684
IHOTarget model=LLaMA-3-8B2026.0684
INPAINTINGTarget model=Qwen-2.5-7B2026.0682
JR1+Target model=LLaMA-3-8B2026.0682
IHOTarget model=CB2026.0682
IHOTarget model=Qwen-2.5-7B+D2026.0680
BONTarget model=LLaMA-3-8B2026.0679
IHOTarget model=CAT2026.0679
INPAINTINGTarget model=Qwen-2.5-32B2026.0678
IHOTarget model=LAT2026.0678
PAIRTarget model=Qwen-2.5-7B2026.0677
PAIR+Target model=LLaMA-3-8B2026.0677
IHOTarget model=CB+D2026.0677
GCG+Target model=LLaMA-3-8B2026.0676
INPAINTINGTarget model=LLaMA-3-8B2026.0675
BONTarget model=Qwen-2.5-7B2026.0673
PAIRTarget model=Qwen-2.5-32B2026.0671
BONTarget model=Qwen-2.5-32B2026.0671
PAIR+Target model=Qwen-2.5-7B+D2026.0671
JR1+Target model=CAT2026.0670
PAIR+Target model=CAT2026.0667
JR1+Target model=Qwen-2.5-7B+D2026.0666
INPAINTINGTarget model=Qwen-2.5-7B+D2026.0666
INPAINTINGTarget model=CB2026.0663
PAIR+Target model=LAT2026.0663
JR1+Target model=LAT2026.0662
INPAINTINGTarget model=LAT2026.0661
PAIR+Target model=CB2026.0658
BONTarget model=Qwen-2.5-7B+D2026.0657
INPAINTINGTarget model=CB+D2026.0655
INPAINTINGTarget model=CAT2026.0654
AmpleGCG+Target model=Qwen-2.5-7B2026.0652
PAIRTarget model=LLaMA-3-8B2026.0649
JR1+Target model=CB2026.0647
PAIR+Target model=CB+D2026.0646
GCG+Target model=LAT2026.0646
BONTarget model=CB2026.0645
PAIRTarget model=Qwen-2.5-7B+D2026.0641
JR1+Target model=CB+D2026.0638
BONTarget model=CB+D2026.0637
PAIRTarget model=CAT2026.0635
AmpleGCG+Target model=Qwen-2.5-32B2026.0631
PAIRTarget model=LAT2026.0631
PAIRTarget model=CB2026.0625
PAIRTarget model=CB+D2026.0617
GCG+Target model=CB2026.0614
AmpleGCG+Target model=CB2026.0611
AmpleGCG+Target model=LLaMA-3-8B2026.068
AmpleGCG+Target model=Qwen-2.5-7B+D2026.066
GCG+Target model=CAT2026.064
AmpleGCG+Target model=CB+D2026.062
BONTarget model=LAT2026.062
BONTarget model=CAT2026.061
AmpleGCG+Target model=LAT2026.060
AmpleGCG+Target model=CAT2026.060