List your company

Multimodal Safety Evaluation on BeaverTails-V

2.9Safety ScoreQwen2.5-VL-7B + TIS
0.49761.12131.7452.3687Mar 3, 2026

Evaluation results

MethodLinks
Qwen2.5-VL-7B + TISBase Model=Qwen2.5-VL-7B, Enhancement=TIS2026.032.92.772.88
Qwen2.5-VL-3B + SFT+GRPOBase Model=Qwen2.5-VL-3B, Stage=SFT+GRPO2026.032.852.752.89
Qwen2.5-VL-3B + SFT+DPO+GRPOBase Model=Qwen2.5-VL-3B, Stage=SFT+DPO+GRPO2026.032.842.782.87
Qwen2.5-VL-7B + SFT+GRPOBase Model=Qwen2.5-VL-7B, Stage=SFT+GRPO2026.032.842.842.91
Qwen2.5-VL-7B + SaFeR-VLMBase Model=Qwen2.5-VL-7B, Method=SaFeR-VLM2026.032.832.332.54
Qwen2.5-VL-7B + SPA-VLBase Model=Qwen2.5-VL-7B, Enhancement=SPA-VL2026.032.822.62.27
Qwen2.5-VL-7B + SFT+DPO+GRPOBase Model=Qwen2.5-VL-7B, Stage=SFT+DPO+GRPO2026.032.822.882.93
Qwen2.5-VL-3B + TISBase Model=Qwen2.5-VL-3B, Enhancement=TIS2026.032.752.372.69
Qwen2.5-VL-3B + SPA-VLBase Model=Qwen2.5-VL-3B, Enhancement=SPA-VL2026.032.742.592.23
Qwen2.5-VL-3B + SaFeR-VLMBase Model=Qwen2.5-VL-3B, Method=SaFeR-VLM2026.032.722.272.28
Qwen2.5-VL-7B + SFT+DPOBase Model=Qwen2.5-VL-7B, Stage=SFT+DPO2026.032.722.812.91
Qwen2.5-VL-7B + VLGuardBase Model=Qwen2.5-VL-7B, Enhancement=VLGuard2026.032.641.731.9
Qwen2.5-VL-7B + SFTBase Model=Qwen2.5-VL-7B, Stage=SFT2026.032.622.662.86
Qwen2.5-VL-3B + VLGuardBase Model=Qwen2.5-VL-3B, Enhancement=VLGuard2026.032.611.691.86
Qwen2.5-VL-3B + SFT+DPOBase Model=Qwen2.5-VL-3B, Stage=SFT+DPO2026.032.582.662.77
Qwen2.5-VL-3B + SFTBase Model=Qwen2.5-VL-3B, Stage=SFT2026.032.572.612.77
Qwen2.5-VL-7B + SIABase Model=Qwen2.5-VL-7B, Enhancement=SIA2026.032.52.332.17
Qwen2.5-VL-7BBase Model=Qwen2.5-VL-7B2026.031.672.72.14
Qwen2.5-VL-7B + ECSOBase Model=Qwen2.5-VL-7B, Enhancement=ECSO2026.031.582.672.15
Qwen2.5-VL-3B + SIABase Model=Qwen2.5-VL-3B, Enhancement=SIA2026.030.822.011.39
Qwen2.5-VL-3B + ECSOBase Model=Qwen2.5-VL-3B, Enhancement=ECSO2026.030.732.411.66
Qwen2.5-VL-3BBase Model=Qwen2.5-VL-3B2026.030.592.561.68