Harmful content detection on ELF22
43.96AccuracyPerspective API
Evaluation results
| Method | Links | |
|---|---|---|
| Perspective API2026.04 | 43.96 | |
| Perspective API2026.04 | 43.96 | |
| OpenAI Moderation2026.04 | 25.85 | |
| OpenAI Moderation2026.04 | 25.85 | |
| LlamaGuard-12026.04 | 15.09 | |
| LlamaGuard-12026.04 | 15.09 | |
| LlamaGuard-22026.04 | 12.07 | |
| LlamaGuard-22026.04 | 12.07 |