List your company

Mathematical Reasoning on AIME 24 (SC Accuracy, Efficiency, Overhead)

80SC AccuracySelf-Consistency
48.48856.66964.8573.031Jan 29, 2026

Evaluation results

MethodLinks
Self-ConsistencyModel=QWQ32B, Sample size N=512026.0180———
PoLRModel=QWQ32B, Sample size N=512026.0180059.710.8
Self-ConsistencyModel=QWQ32B, Sample size N=312026.0178.3———
PoLRModel=QWQ32B, Sample size N=312026.0178.30.761.66.2
Self-ConsistencyModel=QWQ32B, Sample size N=112026.0178.3———
PoLRModel=QWQ32B, Sample size N=112026.0178.3-266.612.8
Self-ConsistencyModel=DSQ7B, Sample size N=112026.0154.3———
PoLRModel=DSQ7B, Sample size N=112026.0154.3-3.349.82
Self-ConsistencyModel=DSQ7B, Sample size N=512026.0153.3———
PoLRModel=DSQ7B, Sample size N=512026.0153.3-6.750.96.3
Self-ConsistencyModel=DSQ7B, Sample size N=312026.0153.3———
PoLRModel=DSQ7B, Sample size N=312026.0153.3-351.53.3
Self-ConsistencyModel=Phi-4-15B, Sample size N=112026.0150.3———
PoLRModel=Phi-4-15B, Sample size N=112026.0150.3-5.358.62.2
Self-ConsistencyModel=Phi-4-15B, Sample size N=512026.0150———
PoLRModel=Phi-4-15B, Sample size N=512026.01503.349.512.1
Self-ConsistencyModel=Phi-4-15B, Sample size N=312026.0149.7———
PoLRModel=Phi-4-15B, Sample size N=312026.0149.70.351.55.3