Bias Detection on Toxigen (test)
90.3AccuracyToxigen
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ToxigenModel Category=Bias Classifier2025.04 | 90.3 | 0.25 | |
| BiasGuard2025.04 | 73.15 | 8 | |
| DeepSeek-R1-32BModel Category=Rule-based LLMs as Judges2025.04 | 70.83 | 13.5 | |
| DeepSeek-R1-32BModel Category=LLMs as Judges2025.04 | 70.3 | 8.85 | |
| GPT-4oModel Category=Rule-based LLMs as Judges2025.04 | 68.35 | 8.45 | |
| GPT-4oModel Category=LLMs as Judges2025.04 | 66.75 | 10.25 | |
| Llama-3-8B-itModel Category=Rule-based LLMs as Judges2025.04 | 63.3 | 12.45 | |
| Moderation APIModel Category=Bias Classifier2025.04 | 60.85 | 0.1 | |
| Azure APIModel Category=Bias Classifier2025.04 | 57.08 | 7.94 | |
| ShieldGemmaModel Category=Bias Classifier2025.04 | 56.2 | 0.3 | |
| Llama-3-8B-itModel Category=LLMs as Judges2025.04 | 50.2 | 24.4 | |
| Llama-Guard-3Model Category=Bias Classifier2025.04 | 49.3 | 9.4 |