RoCode: A Dataset for Measuring Code Intelligence from Problem Definitions in Romanian

Recently, large language models (LLMs) have become increasingly powerful and have become capable of solving a plethora of tasks through proper instructions in natural language. However, the vast majority of testing suites assume that the instructions are written in English, the de facto prompting language. Code intelligence and problem solving still remain a difficult task, even for the most advanced LLMs. Currently, there are no datasets to measure the generalization power for code-generation models in a language other than English. In this work, we present RoCode, a competitive programming dataset, consisting of 2,642 problems written in Romanian, 11k solutions in C, C++ and Python and comprehensive testing suites for each problem. The purpose of RoCode is to provide a benchmark for evaluating the code intelligence of language models trained on Romanian / multilingual text as well as a fine-tuning set for pretrained Romanian models. Through our results and review of related works, we argue for the need to develop code models for languages other than English.

Dataset Details

Dataset Description

• Curated by: Adrian Cosma, Bogdan Iordache, Paolo Rosso
• Language(s) (NLP): Romanian, C++, Python

Dataset Sources

• Repository: https://github.com/cosmaadrian/rocode
• Paper: https://arxiv.org/abs/2402.13222

Uses

Fine-tuning and evaluation of Large-Language Models that solve programming puzzles from Romanian prompts.

Citation

@misc{cosma2024rocode,
      title={RoCode: A Dataset for Measuring Code Intelligence from Problem Definitions in Romanian}, 
      author={Adrian Cosma and Bogdan Iordache and Paolo Rosso},
      year={2024},
      eprint={2402.13222},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Dataset Card Contact

For any information contact Adrian Cosma (cosma.i.adrian@gmail.com)