List your company

Jailbreak Detection on 2-Sigma (test)

90.04AccuracyRF
82.260884.280486.388.3196Feb 10, 2026

Evaluation results

MethodLinks
RFFeatures=LLM-extracted linguistic features2026.0290.0474.6381.9778.120.8712
NBFeatures=LLM-extracted linguistic features2026.0289.6872.8683.6177.860.8749
FDTFeatures=LLM-extracted linguistic features2026.0289.3272.4681.9776.920.8667
LRFeatures=LLM-extracted linguistic features2026.0288.9770.2785.2577.040.8762
XGBoostFeatures=LLM-extracted linguistic features2026.0288.6169.8683.6176.120.868
LGBMFeatures=LLM-extracted linguistic features2026.0287.969.0180.3374.240.8516
DTFeatures=LLM-extracted linguistic features2026.0282.5657.8972.1364.230.7879