Math Reasoning on MiP
1.12FPRRenewal process-based stopping
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Renewal process-based stoppingStopping Rule=Renewal2026.02 | 1.12 | — | — | 71.15 | |
| Maxwise Conformal StoppingStopping Rule=Maxwise2026.02 | 3.21 | — | — | 73.4 | |
| DEERStopping Rule=DEER2026.02 | 16.67 | — | — | 62.34 | |
| EntropyStopping Rule=Entropy2026.02 | 19.39 | — | — | 55.45 | |
| LengthStopping Rule=Length2026.02 | 20.67 | — | — | 61.54 | |
| ConfidenceStopping Rule=Confidence2026.02 | — | -0.48 | 0.48 | — | |
| CriticismStopping Rule=Criticism2026.02 | — | 0.32 | 0.96 | — | |
| No InterventionStopping Rule=No Intervention2026.02 | — | — | 1.12 | — |