Explain-then-Translate: An Analysis on Improving Program Translation with Self-generated Explanations

This work explores the use of self-generated natural language explanations as an intermediate step for code-to-code translation with language models. Across three types of explanations and 19 programming languages constructed from the MultiPL-E dataset, we find the explanations to be particularly ef...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Tang, Zilu, Agarwal, Mayank, Shypula, Alex, Wang, Bailin, Wijaya, Derry, Chen, Jie, Yoon, Kim
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 13.11.2023
Subjects
Datasets
Programming languages
Online AccessGet full text

Cover

More Information
Summary:This work explores the use of self-generated natural language explanations as an intermediate step for code-to-code translation with language models. Across three types of explanations and 19 programming languages constructed from the MultiPL-E dataset, we find the explanations to be particularly effective in the zero-shot case, improving performance by 12% on average. Improvements with natural language explanations are particularly pronounced on difficult programs. We release our dataset, code, and canonical solutions in all 19 languages.
ISSN:2331-8422