List your company

Toxicity Detection on Toxicity Detection 64 model-persona combinations (8 models x 8 personas)

58Win CountSVM Meta-Ensemble
-2.3213.342944.66Jan 5, 2026

Evaluation results

MethodLinks
SVM Meta-EnsembleComparison Baseline=Default2026.0158
Best MajorityComparison Baseline=Default2026.0157
SVM Meta-EnsembleComparison Baseline=Persona Prompt2026.0151
Weighted MajorityComparison Baseline=Default2026.0150
Theoretical Optimal Weighted MajorityComparison Baseline=Default2026.0150
SVM Meta-EnsembleComparison Baseline=Value Profile2026.0147
SVM Meta-EnsembleComparison Baseline=Persona Optimized Prompt2026.0144
Persona Optimized PromptComparison Baseline=Default2026.0143
Value ProfileComparison Baseline=Default2026.0142
SVM Meta-EnsembleComparison Baseline=Theoretical Optimal Weighted Majority2026.0141
Best MajorityComparison Baseline=Persona Prompt2026.0140
SVM Meta-EnsembleComparison Baseline=Weighted Majority2026.0140
Weighted MajorityComparison Baseline=Persona Prompt2026.0137
Theoretical Optimal Weighted MajorityComparison Baseline=Persona Prompt2026.0136
Persona PromptComparison Baseline=Default2026.0134
SVM Meta-EnsembleComparison Baseline=Best Majority2026.0129
Best MajorityComparison Baseline=Persona Optimized Prompt2026.0128
Value ProfileComparison Baseline=Persona Prompt2026.0125
Weighted MajorityComparison Baseline=Value Profile2026.0125
Best MajorityComparison Baseline=Value Profile2026.0124
Persona Optimized PromptComparison Baseline=Persona Prompt2026.0123
Theoretical Optimal Weighted MajorityComparison Baseline=Value Profile2026.0123
Weighted MajorityComparison Baseline=Persona Optimized Prompt2026.0123
Theoretical Optimal Weighted MajorityComparison Baseline=Persona Optimized Prompt2026.0123
Best MajorityComparison Baseline=Theoretical Optimal Weighted Majority2026.0121
Best MajorityComparison Baseline=Weighted Majority2026.0120
Persona Optimized PromptComparison Baseline=Value Profile2026.0119
Value ProfileComparison Baseline=Persona Optimized Prompt2026.0118
Persona PromptComparison Baseline=Value Profile2026.0113
Persona PromptComparison Baseline=Persona Optimized Prompt2026.0112
Value ProfileComparison Baseline=Weighted Majority2026.0110
Value ProfileComparison Baseline=Theoretical Optimal Weighted Majority2026.0110
DefaultComparison Baseline=Value Profile2026.018
DefaultComparison Baseline=Persona Prompt2026.017
Persona Optimized PromptComparison Baseline=Theoretical Optimal Weighted Majority2026.016
DefaultComparison Baseline=Persona Optimized Prompt2026.015
Persona PromptComparison Baseline=Weighted Majority2026.014
Persona PromptComparison Baseline=Theoretical Optimal Weighted Majority2026.014
Persona Optimized PromptComparison Baseline=Weighted Majority2026.014
Weighted MajorityComparison Baseline=Theoretical Optimal Weighted Majority2026.013
Weighted MajorityComparison Baseline=SVM Meta-Ensemble2026.013
Theoretical Optimal Weighted MajorityComparison Baseline=SVM Meta-Ensemble2026.013
Weighted MajorityComparison Baseline=Best Majority2026.012
Best MajorityComparison Baseline=SVM Meta-Ensemble2026.012
Persona PromptComparison Baseline=SVM Meta-Ensemble2026.011
Persona Optimized PromptComparison Baseline=SVM Meta-Ensemble2026.011
Theoretical Optimal Weighted MajorityComparison Baseline=Best Majority2026.011
DefaultComparison Baseline=Weighted Majority2026.010
DefaultComparison Baseline=Best Majority2026.010
DefaultComparison Baseline=Theoretical Optimal Weighted Majority2026.010
DefaultComparison Baseline=SVM Meta-Ensemble2026.010
Persona PromptComparison Baseline=Best Majority2026.010
Value ProfileComparison Baseline=Best Majority2026.010
Value ProfileComparison Baseline=SVM Meta-Ensemble2026.010
Persona Optimized PromptComparison Baseline=Best Majority2026.010
Theoretical Optimal Weighted MajorityComparison Baseline=Weighted Majority2026.010