Code Generation on MBPP+ (F1)
99.32Release RateFirst_p
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| First_palpha=0.202026.05 | 99.32 | 2.72 | 1.415 | |
| Stability2026.05 | 93.24 | 2.9 | 2.4565 | |
| always-valid release wrapperalpha=0.202026.05 | 80.41 | 0 | 4.3025 | |
| Entropy2026.05 | 78.38 | 12.07 | 2.0345 | |
| always-valid release wrapperalpha=0.102026.05 | 77.03 | 0 | 6.2719 | |
| always-valid release wrapperalpha=0.052026.05 | 74.32 | 0.91 | 7.2182 | |
| First_palpha=0.102026.05 | 0 | — | — | |
| First_palpha=0.052026.05 | 0 | — | — |