Harmful content detection on ELF-HP
48.57AccuracyPerspective API
Evaluation results
| Method | Links | |
|---|---|---|
| Perspective API2026.04 | 48.57 | |
| Perspective API2026.04 | 48.57 | |
| OpenAI Moderation2026.04 | 30.63 | |
| OpenAI Moderation2026.04 | 30.63 | |
| LlamaGuard-12026.04 | 21.6 | |
| LlamaGuard-12026.04 | 21.6 | |
| LlamaGuard-22026.04 | 13.94 | |
| LlamaGuard-22026.04 | 13.94 |