Safety Evaluation on Manual Evaluation Set
3.83Average Safety ScoreMSelf-MOA
Evaluation Results
| Method | Links | |
|---|---|---|
| MSelf-MOAModel Family=qwen2026.03 | 3.83 | |
| MPKU-RLHFModel Family=llama2026.03 | 3.8 | |
| MSelf-MOAModel Family=llama2026.03 | 3.67 | |
| MPKU-RLHFModel Family=gemma-32026.03 | 3.67 | |
| MPKU-RLHFModel Family=qwen2026.03 | 3.37 | |
| MSelf-MOAModel Family=gemma-32026.03 | 3.3 | |
| MSelf-MOAModel Family=gemma-22026.03 | 3.2 | |
| MbaseModel Family=gemma-32026.03 | 2.97 | |
| MbaseModel Family=llama2026.03 | 2.77 | |
| MbaseModel Family=qwen2026.03 | 2.43 | |
| MPKU-RLHFModel Family=gemma-22026.03 | 2.13 | |
| MbaseModel Family=gemma-22026.03 | 2.1 |