Code Verification on CLEVER
70Success CountBase Goedel
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Base GoedelPrompting Strategy=Functional, Setting=larger budget2025.11 | 70 | — | 2.5 | |
| QwenPrompting Strategy=Functional, Setting=larger budget2025.11 | 69 | — | 6.3 | |
| Base GoedelPrompting Strategy=Direct, Setting=larger budget2025.11 | 66 | — | — | |
| QwenPrompting Strategy=Direct, Setting=larger budget2025.11 | 59 | — | — | |
| QwenPrompting Strategy=Functional, Setting=pass@12025.11 | 36 | — | 5 | |
| Goedel SFT e30Prompting Strategy=Functional, Setting=larger budget2025.11 | 33 | — | 1.9 | |
| Goedel SFT e30Prompting Strategy=Direct, Setting=larger budget2025.11 | 30 | — | — | |
| QwenPrompting Strategy=Direct, Setting=pass@12025.11 | 28 | — | — |