Safety Evaluation on Manual Evaluation Safety Dataset
3.83Average Safety ScoreM_Self-MOA
Evaluation Results
| Method | Links | |
|---|---|---|
| M_Self-MOABase Model=qwen2026.03 | 3.83 | |
| M_PKU-RLHFBase Model=llama2026.03 | 3.8 | |
| M_Self-MOABase Model=llama2026.03 | 3.67 | |
| M_PKU-RLHFBase Model=gemma-32026.03 | 3.67 | |
| M_PKU-RLHFBase Model=qwen2026.03 | 3.37 | |
| M_Self-MOABase Model=gemma-32026.03 | 3.3 | |
| M_Self-MOABase Model=gemma-22026.03 | 3.2 | |
| M_baseBase Model=gemma-32026.03 | 2.97 | |
| M_baseBase Model=llama2026.03 | 2.77 | |
| M_baseBase Model=qwen2026.03 | 2.43 | |
| M_PKU-RLHFBase Model=gemma-22026.03 | 2.13 | |
| M_baseBase Model=gemma-22026.03 | 2.1 |