List your company

Jailbreak Attack on GPT-J unseen instances

0.61HypervolumeCodeAttack
-0.02440.14030.3050.4697Mar 20, 2026

Evaluation results

MethodLinks
CodeAttackEvaluation Level=ensemble level, Instance Type=unseen instances2026.030.61
EvoJailEvaluation Level=ensemble level, Instance Type=unseen instances2026.030.549
CodeChameleonEvaluation Level=ensemble level, Instance Type=unseen instances2026.030.419
FlipAttackEvaluation Level=ensemble level, Instance Type=unseen instances2026.030.114
ReNeLLMEvaluation Level=ensemble level, Instance Type=unseen instances2026.030.053
JailbrokenEvaluation Level=ensemble level, Instance Type=unseen instances2026.030
CipherEvaluation Level=ensemble level, Instance Type=unseen instances2026.030