List your company

Mathematical Reasoning on MMLU High School Math

94.81Accuracy (Raw consensus)Raw consensus
77.868482.266786.66591.0633May 25, 2026

Evaluation results

MethodLinks
Raw consensusModel=Qwen3-4B, Policy=no-framing majority baseline, Sample count (N)=2702026.0594.81——————
ARBITER-∆Model=Qwen3-4B, Policy=fixed framed+guided parameter-free, Sample count (N)=2702026.0594.8195.190.3710321
Raw consensusModel=Phi-4, Policy=no-framing majority baseline, Sample count (N)=2702026.0592.59——————
ARBITER-∆Model=Phi-4, Policy=fixed framed+guided parameter-free, Sample count (N)=2702026.0592.5993.71.114303
Raw consensusModel=Llama-3.1-8B, Policy=no-framing majority baseline, Sample count (N)=2702026.0578.52——————
ARBITER-∆Model=Llama-3.1-8B, Policy=fixed framed+guided parameter-free, Sample count (N)=2702026.0578.52801.4817954