Math on GSM8K (Solve Rate and Executability)
94.97Solve RateCapFlow
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CapFlowType=Learning, Setting=Multi-domain, Passes=12026.02 | 94.97 | — | |
| ScoreFlowType=Learning, Setting=Single-domain, Iterations=52026.02 | 94.21 | — | |
| AFlowType=Refinement, Setting=Single-domain, Iterations=202026.02 | 93.93 | — | |
| ScoreFlowType=Learning, Setting=Multi-domain, Iterations=52026.02 | 93.45 | — | |
| CapFlowType=Learning, Setting=Cross-domain, Passes=1, Cross-domain Domain Mapping=Reason,Code->Math2026.02 | 92.51 | — | |
| ADASType=Refinement, Setting=Single-domain, Iterations=202026.02 | 92.03 | — | |
| SPPType=Manual, Setting=Manual2026.02 | 91.18 | — | |
| CoT-SCType=Manual, Setting=Manual2026.02 | 90.23 | — | |
| GPT-4o-miniType=Manual, Setting=Manual2026.02 | 90.14 | — | |
| CoTType=Manual, Setting=Manual2026.02 | 89.95 | — | |
| Self-RefineType=Manual, Setting=Manual2026.02 | 88.53 | — |