List your company

Jailbreak Safety Evaluation on Multi-Shot Jailbreak (MSJ)

0ASRST-S
-2.7215.643452.36Jun 15, 2026

Evaluation results

MethodLinks
ST-SBackbone Model=Qwen3-8B2026.060
ST-SBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.060
ST-DBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.060
ST-DBackbone Model=Qwen3-8B2026.062
ST-DBackbone Model=Qwen3-32B2026.062
ST-SBackbone Model=Qwen3-32B2026.064
BaseBackbone Model=Qwen3-8B2026.068
ST-DBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.068
Star1Backbone Model=Qwen3-8B2026.0610
BaseBackbone Model=Qwen3-32B2026.0610
SafePathBackbone Model=Qwen3-32B2026.0610
Star1Backbone Model=DeepSeek-R1-Distill-Llama-70B2026.0610
SafePathBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.0610
Star1Backbone Model=Qwen3-32B2026.0612
ST-SBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.0614
SafePathBackbone Model=Qwen3-8B2026.0618
BaseBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.0618
Star1Backbone Model=DeepSeek-R1-Distill-Llama-8B2026.0654
SafePathBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.0658
BaseBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.0668