LLM Moderation on LLM Moderation (Evaluation Set)
90.23F1 ScoreMLPM
Evaluation results
| Method | Links | |
|---|---|---|
| MLPMModel=OLMo2-7B2025.02 | 90.23 | |
| MLPMModel=Mistral-7B2025.02 | 88.3 | |
| MLPMModel=Llama-8B2025.02 | 87.55 | |
| Prompting(Max)Model=OLMo2-7B, Evaluation Protocol=Prompting (Maximum of 10 prompts)2025.02 | 85.75 | |
| Prompting(Max)Model=Mistral-7B, Evaluation Protocol=Prompting (Maximum of 10 prompts)2025.02 | 85.63 | |
| Prompting(Max)Model=Llama-8B, Evaluation Protocol=Prompting (Maximum of 10 prompts)2025.02 | 82.75 | |
| Prompting(Avg)Model=Llama-8B, Evaluation Protocol=Prompting (Average of 10 prompts)2025.02 | 75.2 | |
| Prompting(Avg)Model=OLMo2-7B, Evaluation Protocol=Prompting (Average of 10 prompts)2025.02 | 75.2 | |
| Prompting(Avg)Model=Mistral-7B, Evaluation Protocol=Prompting (Average of 10 prompts)2025.02 | 72.28 |