List your company

Mathematical Reasoning on MMLU High School

88.1AccuracyComb-E
39.63652.21864.877.382May 18, 2026

Evaluation results

MethodLinks
Comb-EBackbone=GPT-4o-mini, Workflow Type=Combined-E2026.0588.1
IC-QBackbone=GPT-4o-mini, Workflow Type=Learned2026.0588.1
CoTBackbone=GPT-4o-mini, Workflow Type=Chain-of-Thought2026.0586.3
Reflect.Backbone=GPT-4o-mini, Workflow Type=Reflection2026.0585.9
Comb-TBackbone=GPT-4o-mini, Workflow Type=Combined-T2026.0585.2
Direct AnsBackbone=GPT-4o-mini, Workflow Type=Direct2026.0541.5