List your company

Mathematical Reasoning on MMLU Elementary

96AccuracyCoT
63.55271.97680.488.824May 18, 2026

Evaluation results

MethodLinks
CoTBackbone=GPT-4o-mini, Workflow Type=Chain-of-Thought2026.0596
IC-QBackbone=GPT-4o-mini, Workflow Type=Learned2026.0596
Comb-EBackbone=GPT-4o-mini, Workflow Type=Combined-E2026.0595.8
Comb-TBackbone=GPT-4o-mini, Workflow Type=Combined-T2026.0595
Reflect.Backbone=GPT-4o-mini, Workflow Type=Reflection2026.0594.4
Direct AnsBackbone=GPT-4o-mini, Workflow Type=Direct2026.0564.8