Math Reasoning on DeepMath (test)
79.5AccuracyAgentPSO
Evaluation Results
| Method | Links | |
|---|---|---|
| AgentPSOCategory=Multi-Agent2026.05 | 79.5 | |
| DMADCategory=Multi-Agent2026.05 | 72 | |
| Self-RefineCategory=Advanced Single-Agent2026.05 | 70.5 | |
| MoACategory=Multi-Agent2026.05 | 70 | |
| ReflectionCategory=Advanced Single-Agent2026.05 | 69.5 | |
| Self-ConsistencyCategory=Advanced Single-Agent2026.05 | 69.5 | |
| MADCategory=Multi-Agent2026.05 | 69 | |
| Chain-of-ThoughtCategory=Vanilla Single-Agent2026.05 | 68 | |
| Step-Back PromptingCategory=Vanilla Single-Agent2026.05 | 64.5 | |
| ToTCategory=Advanced Single-Agent2026.05 | 63 |