List your company

Math Reasoning on MINERVA (Coverage and Average Success Rate)

39.3CoverageTOFU
27.8630.8333.836.77Apr 30, 2026

Evaluation results

MethodLinks
TOFUBackbone=DEEPSEEK-MATH-7B2026.0439.312
TOFUBackbone=QWEN-2.5-MATH-7B2026.0438.614.7
GEMBackbone=DEEPSEEK-MATH-7B2026.0435.312.5
GEMBackbone=QWEN-2.5-MATH-7B2026.0434.915.1
CEBackbone=DEEPSEEK-MATH-7B2026.0434.212.5
TOFUBackbone=QWEN-2.5-MATH-1.5B2026.0433.510.7
GEMBackbone=QWEN-2.5-MATH-1.5B2026.0431.312.4
CEBackbone=QWEN-2.5-MATH-7B2026.0430.915.1
CEBackbone=QWEN-2.5-MATH-1.5B2026.0428.312.1