List your company

Jailbreak Defense on AdvBench AutoDAN

0ASRDream-v0-Instruct-7B
-1.66929.597920.86532.1321Sep 29, 2025

Evaluation results

MethodLinks
Dream-v0-Instruct-7BBase Model=Dream-v0-Instruct-7B, Defense Method=None2025.090
Dream-v0-Instruct-7B + PPL-FilterBase Model=Dream-v0-Instruct-7B, Defense Method=PPL-Filter2025.090
Dream-v0-Instruct-7B + DIFFUGUARDBase Model=Dream-v0-Instruct-7B, Defense Method=DIFFUGUARD2025.090
Dream-v0-Instruct-7B + Self-reminderBase Model=Dream-v0-Instruct-7B, Defense Method=Self-reminder2025.090
Dream-v0-Instruct-7B + Self-reminder + DIFFUGUARDBase Model=Dream-v0-Instruct-7B, Defense Method=Self-reminder + DIFFUGUARD2025.090
MMaDA-8B-MixCoT + Self-reminder + DIFFUGUARDBase Model=MMaDA-8B-MixCoT, Defense Method=Self-reminder + DIFFUGUARD2025.095.77
MMaDA-8B-MixCoT + DIFFUGUARDBase Model=MMaDA-8B-MixCoT, Defense Method=DIFFUGUARD2025.096.35
MMaDA-8B-MixCoT + Self-reminderBase Model=MMaDA-8B-MixCoT, Defense Method=Self-reminder2025.0914.23
LLaDA-8B-Instruct + Self-reminder + DIFFUGUARDBase Model=LLaDA-8B-Instruct, Defense Method=Self-reminder + DIFFUGUARD2025.0916.73
LLaDA-8B-Instruct + Self-reminderBase Model=LLaDA-8B-Instruct, Defense Method=Self-reminder2025.0920.77
LLaDA-1.5 + Self-reminder + DIFFUGUARDBase Model=LLaDA-1.5, Defense Method=Self-reminder + DIFFUGUARD2025.0920.96
MMaDA-8B-MixCoT + PPL-FilterBase Model=MMaDA-8B-MixCoT, Defense Method=PPL-Filter2025.0921.54
MMaDA-8B-MixCoTBase Model=MMaDA-8B-MixCoT, Defense Method=None2025.0924.81
LLaDA-1.5 + Self-reminderBase Model=LLaDA-1.5, Defense Method=Self-reminder2025.0926.15
LLaDA-8B-Instruct + DIFFUGUARDBase Model=LLaDA-8B-Instruct, Defense Method=DIFFUGUARD2025.0931.54
LLaDA-8B-Instruct + PPL-FilterBase Model=LLaDA-8B-Instruct, Defense Method=PPL-Filter2025.0934.23
LLaDA-1.5 + DIFFUGUARDBase Model=LLaDA-1.5, Defense Method=DIFFUGUARD2025.0936.92
LLaDA-1.5 + PPL-FilterBase Model=LLaDA-1.5, Defense Method=PPL-Filter2025.0938.08
LLaDA-8B-InstructBase Model=LLaDA-8B-Instruct, Defense Method=None2025.0939.23
LLaDA-1.5Base Model=LLaDA-1.5, Defense Method=None2025.0941.73