List your company

Safety Evaluation on Manual Evaluation Safety Dataset

3.83Average Safety ScoreM_Self-MOA
2.03082.49792.9653.4321Mar 7, 2026

Evaluation results

MethodLinks
M_Self-MOABase Model=qwen2026.033.83
M_PKU-RLHFBase Model=llama2026.033.8
M_Self-MOABase Model=llama2026.033.67
M_PKU-RLHFBase Model=gemma-32026.033.67
M_PKU-RLHFBase Model=qwen2026.033.37
M_Self-MOABase Model=gemma-32026.033.3
M_Self-MOABase Model=gemma-22026.033.2
M_baseBase Model=gemma-32026.032.97
M_baseBase Model=llama2026.032.77
M_baseBase Model=qwen2026.032.43
M_PKU-RLHFBase Model=gemma-22026.032.13
M_baseBase Model=gemma-22026.032.1