Bias Detection on The Gab Hate (GHC) (test)
71.25AccuracyBiasGuard
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BiasGuard2025.04 | 71.25 | 12.5 | |
| GPT-4oModel Category=Rule-based LLMs as Judges2025.04 | 70.94 | 16.5 | |
| Llama-3-8B-itModel Category=Rule-based LLMs as Judges2025.04 | 68.15 | 10.05 | |
| DeepSeek-R1-32BModel Category=Rule-based LLMs as Judges2025.04 | 67.95 | 11.15 | |
| GPT-4oModel Category=LLMs as Judges2025.04 | 62.1 | 16.05 | |
| ToxigenModel Category=Bias Classifier2025.04 | 60.25 | 4.85 | |
| Moderation APIModel Category=Bias Classifier2025.04 | 60.25 | 0.65 | |
| Llama-3-8B-itModel Category=LLMs as Judges2025.04 | 55.25 | 18.25 | |
| Azure APIModel Category=Bias Classifier2025.04 | 49.25 | 25.13 | |
| Llama-Guard-3Model Category=Bias Classifier2025.04 | 49.05 | 2.65 | |
| DeepSeek-R1-32BModel Category=LLMs as Judges2025.04 | 47.2 | 27.25 | |
| ShieldGemmaModel Category=Bias Classifier2025.04 | 22.95 | 2.05 |