Hate speech classification on Stormfront (test)
0.921AUCxList-Hate
Evaluation results
| Method | Links | ||
|---|---|---|---|
| xList-HateModel=mistral-24b2026.02 | 0.921 | 0.106 | |
| xList-HateModel=mistral-7b2026.02 | 0.877 | 0.076 | |
| xList-HateModel=gemma3-27b2026.02 | 0.877 | 0.071 | |
| xList-HateModel=gemma3-12b2026.02 | 0.866 | 0.054 | |
| xList-HateModel=gemma3-4b2026.02 | 0.86 | 0.064 | |
| xList-HateModel=llama3-3b2026.02 | 0.85 | 0.084 | |
| Zero-shotModel=mistral-24b2026.02 | 0.815 | — | |
| Zero-shotModel=gemma3-12b2026.02 | 0.813 | — | |
| Zero-shotModel=gemma3-27b2026.02 | 0.806 | — | |
| Zero-shotModel=mistral-7b2026.02 | 0.8 | — | |
| Zero-shotModel=gemma3-4b2026.02 | 0.797 | — | |
| Zero-shotModel=llama3-8b2026.02 | 0.795 | — | |
| xList-HateModel=llama3-8b2026.02 | 0.784 | -0.011 | |
| Zero-shotModel=llama3-3b2026.02 | 0.765 | — | |
| Zero-shotModel=llama3-1b2026.02 | 0.726 | — | |
| xList-HateModel=llama3-1b2026.02 | 0.708 | -0.018 |