| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| HarmfulQA | SafeMoE-XL | Safety Rate98.1 | 15 | 1d ago | |
| PKU-Safe | SafeMoE-XL | Safety Rate90.83 | 11 | 1d ago | |
| BeaverTails | SafeMoE-XL | Safety Rate87.2 | 11 | 1d ago | |
| HarmBench | SafeMoE-XL | Safety Rate82.5 | 11 | 1d ago | |
| AdvBench | SafeMoE-XL | Safety Rate97.2 | 11 | 1d ago |