Toxicity Detection on LMSYS-Chat-1M
0.9669AccuracyMULI
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MULI2024.05 | 0.9669 | 0.9823 | 0.985 | 0.8865 | 0.6685 | 0.5362 | |
| OModfull_name=OpenAI Moderation API2024.05 | 0.9597 | 0.9762 | 0.9816 | 0.8159 | 0.6374 | 0.5695 | |
| LlamaGuard2024.05 | 0.9389 | 0.9272 | 0.9344 | 0.6752 | 0.0729 | 0.0028 | |
| Logits_Concat2024.05 | 0.8964 | 0.836 | 0.8209 | 0.4366 | 0.02 | 0 |