Toxicity Detection on ToxicChat (test)
0.9772AccuracyMULI
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MULI2024.05 | 0.9772 | 0.9129 | 0.9834 | 0.8122 | 0.4254 | 0.2403 | |
| LlamaGuard2024.05 | 0.9553 | 0.7014 | 0.9088 | 0.4972 | 0.0525 | 0.0138 | |
| OMod2024.05 | 0.9494 | 0.6314 | 0.8619 | 0.3895 | 0.0608 | 0.0276 | |
| Logits_Cannot2024.05 | 0.9457 | 0.5401 | 0.7072 | 0.3398 | 0.0829 | 0.0552 |