List your company

Semantic Hijacking Attack Success Evaluation on Semantic Hijacking Evaluation Set

94.4ASRLlama-3.1-8B
-2.94422.32847.672.872May 17, 2026

Evaluation results

MethodLinks
Llama-3.1-8BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0594.4———
Mistral-Small-3.2-24BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0591———
Qwen-2.5-72BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0591———
Qwen-2.5-7BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0589.4———
Qwen-3-32BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0582.4———
Qwen-2.5-7BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0579.6———
DeepSeek-R1Role=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0576.4———
Llama-3.1-8BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0575.8———
Qwen-2.5-72BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0574.4———
DeepSeek-R1Role=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0574.4———
Mistral-NemoRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0572.2———
Llama-3.1-70BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0571———
Mistral-Small-3.2-24BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0566———
Llama-3.1-8BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0565.4———
Llama-3.1-70BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0565.2———
GPT-4o-MiniRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0564.8———
Qwen-2.5-7BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0562———
Mistral-Small-3.2-24BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0560———
Qwen-2.5-7BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0559.2———
Mistral-Small-3.2-24BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0558.6———
Qwen-3-32BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0558.4———
Qwen-2.5-72BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0556.4———
Mistral-NemoRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0551.2———
GPT-4o-MiniRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0550.4———
Qwen-2.5-72BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0549.6———
Qwen-2.5-7BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0549———
Llama-3.1-70BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0547.6———
Mistral-Small-3.2-24BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0547.4———
Llama-3.1-70BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0544.4———
Qwen-2.5-7BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0543.8———
Mistral-Small-3.2-24BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0543.4———
Llama-3.1-8BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0542.2———
Qwen-3-32BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0541.2———
Mistral-NemoRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0540.8———
Llama-3.1-8BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0540.6———
Llama-3.1-8BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0538.2———
Qwen-3.5-27BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0537.2———
Qwen-2.5-7BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0533.8———
Mistral-Small-3.2-24BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0533.6———
GPT-4o-MiniRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0531.8———
Qwen-3-32BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0531.8———
Qwen-2.5-72BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0531———
Llama-3.1-70BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0530.2———
Qwen-2.5-72BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0529———
Gemma-4-31BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0526.6———
DeepSeek-R1Role=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0525.6———
Qwen-2.5-72BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0523.4———
Mistral-NemoRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0522.2———
GPT-4o-MiniRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0522———
Llama-3.1-70BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0521.6———
DeepSeek-R1Role=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0521.6———
Llama-3.1-8BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0520.6———
Llama-3.1-70BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0520.2———
DeepSeek-R1Role=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0520.2———
DeepSeek-R1Role=Manager, Worker Model=GPT-4o, Configuration=Config A2026.0519.4———
Gemma-4-31BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0517———
GPT-4o-MiniRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0515.6———
GPT-4o-MiniRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0515.6———
Qwen-3.5-9BRole=Manager, Worker Model=DeepSeek-R1, Configuration=Config A2026.0514.6———
Qwen-3-32BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.0514.4———
Qwen-3.5-27BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0514———
Qwen-3-32BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0513.8———
Mistral-NemoRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.0513———
Qwen-3.5-9BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0512.6———
Gemma-4-31BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0512———
Qwen-3.5-27BRole=Manager, Worker Model=Llama-3.1-70B, Configuration=Config A2026.0511.6———
Mistral-NemoRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.0511.4———
Qwen-3.5-9BRole=Manager, Worker Model=Qwen-2.5-7B, Configuration=Config A2026.0511.2———
Qwen-3-32BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.058.4———
Qwen-3.5-27BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.056.4———
Qwen-3.5-9BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.056.4———
GPT-4o-MiniRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.056———
Qwen-3.5-27BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.056———
Qwen-3.5-9BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.055.8———
Gemma-4-31BRole=Manager, Worker Model=Qwen-2.5-72B, Configuration=Config A2026.055.6———
Gemma-4-31BRole=Manager, Worker Model=GPT-4o, Configuration=Config A2026.055———
Mistral-NemoRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.054.4———
DeepSeek-R1Role=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.053.8———
Qwen-3.5-27BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.053———
Gemma-4-31BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.052———
Qwen-3.5-9BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.051.8———
Qwen-3.5-27BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.051.6———
Qwen-3.5-9BRole=Manager, Worker Model=Mistral-Small-3.2-24B, Configuration=Config A2026.051.4———
Gemma-4-31BRole=Manager, Worker Model=Llama-3.1-8B, Configuration=Config A2026.050.8———
DeepSeek-R1Role=Manager2026.05—34.49——
DeepSeek-R1Role=Worker, Metric Source=Average Row2026.05—63.85——
DeepSeek-R1Role=Worker, Configuration=Config B2026.05——90.3—
DeepSeek-R1Role=Worker, Configuration=Config D2026.05——81.5—
DeepSeek-R1Role=Worker, Configuration=Delta2026.05———-8.8
Gemma-4-31BRole=Manager2026.05—9.86——
GPT-4oRole=Worker, Metric Source=Average Row2026.05—31.75——
GPT-4oRole=Worker, Configuration=Config B2026.05——29.2—
GPT-4oRole=Worker, Configuration=Config D2026.05——90.4—
GPT-4oRole=Worker, Configuration=Delta2026.05———61.2
GPT-4o-MiniRole=Manager2026.05—29.46——
Llama-3.1-70BRole=Manager2026.05—42.89——
Llama-3.1-70BRole=Worker, Metric Source=Average Row2026.05—39.52——
Llama-3.1-70BRole=Worker, Configuration=Config B2026.05——51.3—
Llama-3.1-70BRole=Worker, Configuration=Config D2026.05——98—
Llama-3.1-70BRole=Worker, Configuration=Delta2026.05———46.7