List your company

Safety Evaluation on WildGuardMix

0.8974Safety ScoreImplicitRM
0.3603440.4997720.63920.778628Mar 19, 2026Mar 20, 2026Mar 21, 2026Mar 22, 2026Mar 23, 2026Mar 24, 2026

Evaluation results

MethodLinks
ImplicitRMPolicy model=Qwen3-Instruct-4B2026.030.897410.9
ImplicitRMPolicy model=Qwen2.5-Instruct-7B2026.030.882715.3
LAGAMPolicy model=Qwen3-Instruct-4B2026.030.87017.5
SelectMixPolicy model=Qwen3-Instruct-4B2026.030.86727.2
Causal-DRPolicy Model=LLaMA2-7B2026.030.86511.3
SDR2Policy model=Qwen3-Instruct-4B2026.030.85345.4
Causal-IPSPolicy Model=LLaMA2-7B2026.030.8428.3
SelectMixPolicy model=Qwen2.5-Instruct-7B2026.030.84129.9
LAGAMPolicy model=Qwen2.5-Instruct-7B2026.030.83669.3
SDRPolicy Model=LLaMA2-7B2026.030.836.8
SDR2Policy model=Qwen2.5-Instruct-7B2026.030.81826.9
ILDEPolicy Model=LLaMA2-7B2026.030.8154.9
NaivePolicy model=Qwen3-Instruct-4B2026.030.8094—
IPSPolicy Model=LLaMA2-7B2026.030.8094.2
NaivePolicy Model=LLaMA2-7B2026.030.777—
NaivePolicy model=Qwen2.5-Instruct-7B2026.030.7654—
Causal-DRPolicy Model=Qwen2.5-7B2026.030.56949.2
Causal-IPSPolicy Model=Qwen2.5-7B2026.030.55144.6
SDRPolicy Model=Qwen2.5-7B2026.030.53740.8
ILDEPolicy Model=Qwen2.5-7B2026.030.53440
IPSPolicy Model=Qwen2.5-7B2026.030.51635.4
NaivePolicy Model=Qwen2.5-7B2026.030.381—