Bootstrapping Niche Multilingual Code Translation via Reinforcement Learning with Execution-Based Verifiable Supervision

• URL: http://arxiv.org/abs/2608.13854v1
• Date: Fri, 14 Aug 2026 00:56:41 GMT
• Title: Bootstrapping Niche Multilingual Code Translation via Reinforcement Learning with Execution-Based Verifiable Supervision
• Abstract summary: We introduce HumanEval-X++, an execution-based benchmark that extends HumanEval-X to a broad many-to-many language space.<n>On HumanEval-X++ and existing benchmarks, it yields consistent gains over the untrained baselines.<n>Our study establishes a reliable approach of data generation, training, and benchmarking, paving the way toward further bootstrapping the quality of many-to-many translation for programming languages.
• Score: 41.68344013253776
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Code translation must preserve executable behavior across many programming languages, yet neural code translation has largely focused on a few popular languages such as C++, Java, and Python. This leaves a niche, many-to-many setting where parallel supervision is sparse, producing plausible but non-executable translations. We address this setting with preference-based reinforcement learning driven by execution-based supervision. Our pipeline firstly expands verifiable seed Python programs into a multilingual pool of execution-validated codes. Using the pool, a base LLM generates translation candidates across language pairs, which we label by their execution outcomes. The resulting preferences are used to train a reward model that scores cross-language translation quality. Finally, we optimize our base LLMs with GRPO over 600 directed language pairs (25 x 24) using the reward model as a signal. To evaluate the niche translation capability, we introduce HumanEval-X++, an execution-based benchmark that extends HumanEval-X to a broad many-to-many language space. We evaluate our approach using Qwen-3.5 4B and 9B models. On HumanEval-X++ and existing benchmarks, it yields consistent gains over the untrained baselines. In particular, the 4B model achieves an average improvement of 13% across all languages on HumanEval-X++, with a gain of 21% on mid-tier languages. Our study establishes a reliable approach of data generation, training, and benchmarking, paving the way toward further bootstrapping the quality of many-to-many translation for programming languages.

Related papers

• Scaling Laws for Code: Every Programming Language Matters [73.6302896079007]
  Code large language models (Code LLMs) are powerful but costly to train.<n>Different programming languages (PLs) have varying impacts during pre-training that significantly affect base model performance.<n>We present the first systematic exploration of scaling laws for multilingual code pre-training.
  arXiv  Detail & Related papers  (2025-12-15T16:07:34Z)
• A Multi-Language Object-Oriented Programming Benchmark for Large Language Models [61.267115598083315]
  A survey of 35 existing benchmarks uncovers three major imbalances.<n>85.7% focus on a single programming language.<n>94.3% target only function-level or statement-level tasks.<n>Over 80% include fewer than ten test cases on average.
  arXiv  Detail & Related papers  (2025-09-30T11:30:08Z)
• Unraveling the Potential of Large Language Models in Code Translation: How Far Are We? [4.616570111453259]
  Large language models (LLMs) exhibit state-of-the-art performance in various tasks, but struggle for code translation. We conduct a large-scale empirical study to exploit the capabilities and incapabilities of LLMs in code translation tasks. We propose two methods: (1) intermediary translation which selects an intermediary language between the source and target ones; and (2) self-training which fine-tunes LLMs on self-generated parallel data.
  arXiv  Detail & Related papers  (2024-10-13T12:20:12Z)
• CRUXEval-X: A Benchmark for Multilingual Code Reasoning, Understanding and Execution [50.1875460416205]
  The CRUXEVAL-X code reasoning benchmark contains 19 programming languages.<n>It comprises at least 600 subjects for each language, along with 19K content-consistent tests in total.<n>Even a model trained solely on Python can achieve at most 34.4% Pass@1 in other languages.
  arXiv  Detail & Related papers  (2024-08-23T11:43:00Z)
• Breaking Language Barriers in Multilingual Mathematical Reasoning: Insights and Observations [59.056367787688146]
  This paper pioneers exploring and training powerful Multilingual Math Reasoning (xMR) LLMs. We construct the first multilingual math reasoning instruction dataset, MGSM8KInstruct, encompassing ten distinct languages. By utilizing translation, we construct the first multilingual math reasoning instruction dataset, MGSM8KInstruct, encompassing ten distinct languages.
  arXiv  Detail & Related papers  (2023-10-31T08:09:20Z)
• Measuring The Impact Of Programming Language Distribution [28.96076723773365]
  We present the BabelCode framework for execution-based evaluation of any benchmark in any language. We present a new code translation dataset called Translating Python Programming Puzzles (TP3) We investigate if balancing the distributions of 14 languages in a training dataset improves a large language model's performance on low-resource languages.
  arXiv  Detail & Related papers  (2023-02-03T19:47:22Z)
• XNLI 2.0: Improving XNLI dataset and performance on Cross Lingual Understanding (XLU) [0.0]
  We focus on improving the original XNLI dataset by re-translating the MNLI dataset in all of the 14 different languages present in XNLI. We also perform experiments by training models in all 15 languages and analyzing their performance on the task of natural language inference.
  arXiv  Detail & Related papers  (2023-01-16T17:24:57Z)
• Multi-lingual Evaluation of Code Generation Models [82.7357812992118]
  We present new benchmarks on evaluation code generation models: MBXP and Multilingual HumanEval, and MathQA-X. These datasets cover over 10 programming languages. We are able to assess the performance of code generation models in a multi-lingual fashion.
  arXiv  Detail & Related papers  (2022-10-26T17:17:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.