From Reasoning to Code: GRPO Optimization for Underrepresented Languages
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911714763603968 |
|---|---|
| author | Pennino, Federico Raimondi, Bianca Rondelli, Massimo Gurioli, Andrea Gabbrielli, Maurizio |
| author_facet | Pennino, Federico Raimondi, Bianca Rondelli, Massimo Gurioli, Andrea Gabbrielli, Maurizio |
| contents | Generating accurate and executable code using Large Language Models (LLMs) remains a significant challenge for underrepresented programming languages, such as Prolog and Lisp, due to the scarcity of public training data compared to high-resource languages like Python. This paper introduces a generalizable Reinforcement Learning (RL) approach that combines small-scale versions of the Qwen2.5-Coder model with Group Relative Policy Optimization (GRPO) to enable effective code generation through reasoning. To address the limitations of sparse datasets, we integrate execution-driven feedback directly into the RL loop, utilizing a reward system that exploits both logical correctness and structural formatting. Experimental results on GSM8K dataset demonstrate significant improvements in reasoning quality and code accuracy across underrepresented languages. These findings underscore the potential of our approach to benefit a wide range of programming languages lacking extensive training resources by leveraging symbolic reasoning and interpreter-based feedback. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_11027 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Reasoning to Code: GRPO Optimization for Underrepresented Languages Pennino, Federico Raimondi, Bianca Rondelli, Massimo Gurioli, Andrea Gabbrielli, Maurizio Machine Learning Artificial Intelligence Programming Languages Generating accurate and executable code using Large Language Models (LLMs) remains a significant challenge for underrepresented programming languages, such as Prolog and Lisp, due to the scarcity of public training data compared to high-resource languages like Python. This paper introduces a generalizable Reinforcement Learning (RL) approach that combines small-scale versions of the Qwen2.5-Coder model with Group Relative Policy Optimization (GRPO) to enable effective code generation through reasoning. To address the limitations of sparse datasets, we integrate execution-driven feedback directly into the RL loop, utilizing a reward system that exploits both logical correctness and structural formatting. Experimental results on GSM8K dataset demonstrate significant improvements in reasoning quality and code accuracy across underrepresented languages. These findings underscore the potential of our approach to benefit a wide range of programming languages lacking extensive training resources by leveraging symbolic reasoning and interpreter-based feedback. |
| title | From Reasoning to Code: GRPO Optimization for Underrepresented Languages |
| topic | Machine Learning Artificial Intelligence Programming Languages |
| url | https://arxiv.org/abs/2506.11027 |