Natural Language to Code Translation with Execution

Benchmark Model Rank Results
code-generation-on-mbppcode-davinci-001 175B + MBR-Exec#39Accuracy: 58.2