Multi-Lingual Evaluation of Code Generation Models
Abstract
We present two new benchmarks, MBXP and Multilingual HumanEval, designed to evaluate code completion models in over 10 programming languages. These datasets are generated using a conversion framework that transpiles prompts and test cases from the original MBPP and HumanEval datasets into the corresponding data in the target language. By using these benchmarks, we are able to assess the performance of code generation models in a multi-lingual fashion, and discovered generalization ability of language models on out-of-domain languages, advantages of multi-lingual models over mono-lingual, the ability of few-shot prompting to teach the model new languages, and zero-shot translation abilities. In addition, we use our code generation model to perform large-scale bootstrapping to obtain synthetic canonical solutions in several languages, which can be used for other code-related evaluations such as code insertion, robustness, or summarization tasks.
Cite
Text
Athiwaratkun et al. "Multi-Lingual Evaluation of Code Generation Models." International Conference on Learning Representations, 2023.Markdown
[Athiwaratkun et al. "Multi-Lingual Evaluation of Code Generation Models." International Conference on Learning Representations, 2023.](https://mlanthology.org/iclr/2023/athiwaratkun2023iclr-multilingual/)BibTeX
@inproceedings{athiwaratkun2023iclr-multilingual,
title = {{Multi-Lingual Evaluation of Code Generation Models}},
author = {Athiwaratkun, Ben and Gouda, Sanjay Krishna and Wang, Zijian and Li, Xiaopeng and Tian, Yuchen and Tan, Ming and Ahmad, Wasi Uddin and Wang, Shiqi and Sun, Qing and Shang, Mingyue and Gonugondla, Sujan Kumar and Ding, Hantian and Kumar, Varun and Fulton, Nathan and Farahani, Arash and Jain, Siddhartha and Giaquinto, Robert and Qian, Haifeng and Ramanathan, Murali Krishna and Nallapati, Ramesh and Ray, Baishakhi and Bhatia, Parminder and Sengupta, Sudipta and Roth, Dan and Xiang, Bing},
booktitle = {International Conference on Learning Representations},
year = {2023},
url = {https://mlanthology.org/iclr/2023/athiwaratkun2023iclr-multilingual/}
}