| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| BeaverTails | Roblox Guard 1.0 | F1 Score87.3 | 60 | 1mo ago | |
| WildGuard | SingGuard-2B | F1 Score80.26 | 26 | 1mo ago | |
| OpenAI | GPT-4o | F1 Score82.3 | 26 | 4mo ago | |
| Toxic Chat | GGuard | F1 Score82 | 24 | 4mo ago | |
| Aegis | OMNIGUARD-7B | F1 Score84 | 12 | 4mo ago |