List your company

Math Reasoning on DeepMath (test)

79.5AccuracyAgentPSO
62.3466.79571.2575.705May 9, 2026

Evaluation results

MethodLinks
AgentPSOCategory=Multi-Agent2026.0579.5
DMADCategory=Multi-Agent2026.0572
Self-RefineCategory=Advanced Single-Agent2026.0570.5
MoACategory=Multi-Agent2026.0570
ReflectionCategory=Advanced Single-Agent2026.0569.5
Self-ConsistencyCategory=Advanced Single-Agent2026.0569.5
MADCategory=Multi-Agent2026.0569
Chain-of-ThoughtCategory=Vanilla Single-Agent2026.0568
Step-Back PromptingCategory=Vanilla Single-Agent2026.0564.5
ToTCategory=Advanced Single-Agent2026.0563