Harmful content detection on COVID-HATE
96.4AccuracyPerspective API
Evaluation Results
| Method | Links | |
|---|---|---|
| Perspective API2026.04 | 96.4 | |
| Perspective API2026.04 | 96.4 | |
| OpenAI Moderation2026.04 | 87.89 | |
| OpenAI Moderation2026.04 | 87.89 | |
| LlamaGuard-12026.04 | 58.56 | |
| LlamaGuard-12026.04 | 58.56 | |
| LlamaGuard-22026.04 | 34.83 | |
| LlamaGuard-22026.04 | 34.83 |