Code Generation on MBPP (MA, MI, Error Rate)
0.68MA
PROBELLM
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.02 | 0.68 | 0.32 | 0.32 | |
2026.02 | 0.65 | 0.35 | 0.46 | |
2026.02 | 0.59 | 0.41 | 0.76 | |
2026.02 | 0.53 | 0.47 | 0.54 | |
2026.02 | 0.5 | 0.5 | 0.83 | |
2026.02 | 0.42 | 0.58 | 0.62 | |
2026.02 | 0.42 | 0.58 | 0.64 | |
2026.02 | 0.4 | 0.6 | 0.9 | |
2026.02 | 0.39 | 0.61 | 0.78 | |
2026.02 | 0.35 | 0.65 | 0.39 | |
2026.02 | 0.31 | 0.69 | 0.76 | |
2026.02 | 0.27 | 0.73 | 0.5 |