Harmful content detection on CONAN
98.47AccuracyLlamaGuard-1
Evaluation Results
| Method | Links | |
|---|---|---|
| LlamaGuard-12026.04 | 98.47 | |
| LlamaGuard-12026.04 | 98.47 | |
| Perspective API2026.04 | 96.97 | |
| Perspective API2026.04 | 96.97 | |
| OpenAI Moderation2026.04 | 95.29 | |
| OpenAI Moderation2026.04 | 95.29 | |
| LlamaGuard-22026.04 | 86.65 | |
| LlamaGuard-22026.04 | 86.65 |