Mathematical Reasoning on Game of 24 (pass@k)
100pass@1TSLM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TSLMevaluation_protocol=pass@12026.01 | 100 | — | |
| Procedure Cloningevaluation_protocol=pass@12026.01 | 47 | — | |
| SCevaluation_protocol=pass@12026.01 | 17 | — | |
| GRPOevaluation_protocol=pass@12026.01 | 15 | — | |
| Tree-of-Thoughtevaluation_protocol=pass@1002026.01 | — | 32 |