List your company

Semantic Hijacking on Semantic Hijacking Mutation Pipeline Config B

0FR RateMistral-Nemo
-3.61220.76945.1569.531May 17, 2026

Evaluation results

MethodLinks
Mistral-NemoEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0504674670033
Llama-3.1-8BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.052.84984840142
Gemma-3-12BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.057.45004630370
Llama-3.3-70BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0510.44994470521
Mixtral-8x22BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0510.85004460540
Gemma-2-27BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0513.34974310663
Qwen-2.5-72BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0513.44994320671
Mistral-Small-3.2-24BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0524.249937621211
Qwen-3.5-9BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0527.6468338112932
GPT-4o-MiniEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0529.250035401460
Llama-3.1-70BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0551.349924302561
Gemma-4-31BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0561.149619303034
Qwen-3.5-27BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0564.249517703185
Qwen-2.5-7BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0570.849783623523
Qwen-3-32BEvaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0578.2490107038310
DeepSeek-V3Evaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0589.45005304470
DeepSeek-R1Evaluation Mode=Worker-Only, Configuration=Config B, Sample Size=n=5002026.0590.34964804484