List your company

Toxicity Mitigation on RealToxicityPrompts 1k samples

0.51CLS ToxicityPID-AcT
0.29841.72673.1554.5833Oct 5, 2025

Evaluation results

MethodLinks
PID-AcTModel=Gemma2-2B, Sequential=true2025.100.512.93.4
Mean-AcTModel=Gemma2-2B, Sequential=true2025.100.683.233.7
PID-AcTModel=Llama3-8B, Sequential=true2025.100.724.364.9
ITI-CModel=Gemma2-2B, Sequential=false2025.100.745.366.1
Linear-AcTModel=Gemma2-2B, Sequential=false2025.100.955.375.92
Linear-AcTModel=Gemma2-2B, Sequential=true2025.1014.134.64
Mean-AcTModel=Gemma2-2B, Sequential=false2025.101.125.25.8
CAAModel=Gemma2-2B, Sequential=false2025.101.25.355.88
Mean-AcTModel=Llama3-8B, Sequential=true2025.101.215.095.73
ITI-CModel=Llama3-8B, Sequential=false2025.101.66.537.19
Linear-AcTModel=Llama3-8B, Sequential=true2025.101.686.477.12
Mean-AcTModel=Llama3-8B, Sequential=false2025.101.786.567.3
CAAModel=Llama3-8B, Sequential=false2025.101.826.77.4
Linear-AcTModel=Llama3-8B, Sequential=false2025.101.876.557.3
AURAModel=Llama3-8B, Sequential=false2025.101.98.128.8
AURAModel=Gemma2-2B, Sequential=false2025.102.129.049.72
ActADDModel=Gemma2-2B, Sequential=false2025.103.9613.4314.17
OriginalModel=Gemma2-2B, Sequential=false2025.104.1713.4214.17
ActADDModel=Llama3-8B, Sequential=false2025.105.5715.7316.48
OriginalModel=Llama3-8B, Sequential=false2025.105.81515.81