Metraipolis › archive › dc_06e61a06ee90

CodeTransBenchmark: Evaluating LLM-based Code Translation and Repair Across Programming Languages — arXiv:2609.20257

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.20257
License (as archived)
CC0-1.0
Archived
2026-10-10T04:15:20.062Z
SHA-256 of the archived text
1006780568a89b3f2d71f0ae9b82352ae1e4f629448f0f01e2fe6db8231d1f64
Stable URL of this page
https://metraipolis.com/documento/dc_06e61a06ee90

How to cite

CodeTransBenchmark: Evaluating LLM-based Code Translation and Repair Across Programming Languages — arXiv:2609.20257 — arXiv (en), https://arxiv.org/abs/2609.20257 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_06e61a06ee90 (Metraipolis archive dc_06e61a06ee90, sha256 1006780568a89b3f2d71f0ae9b82352ae1e4f629448f0f01e2fe6db8231d1f64)
BibTeX
@misc{dc06e61a06ee90,
  title = {CodeTransBenchmark: Evaluating LLM-based Code Translation and Repair Across Programming Languages — arXiv:2609.20257},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20257},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_06e61a06ee90; sha256 1006780568a89b3f2d71f0ae9b82352ae1e4f629448f0f01e2fe6db8231d1f64},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20257. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Models (LLMs) pre-trained on expansive text and code corpora have revealed promising code generation abilities and have attracted increasing attention in code translation. In this work, we investigate the effectiveness of LLMs in code translation and translation error repair. First, we present CodeTransBenchmark, a framework for evaluating LLM-based translation and repair and devise a post-processing strategy to extract code from inconsistent LLM outputs. Then, we discuss an empirical study evaluates eight models on three datasets and 12 language pairs, in which we categorize incorrect translations by errors to identify weaknesses of existing LLMs. Our work shows that while LLMs specifically trained for multi-lingual coding, like Codestral, correctly translate the majority of code, most general-purpose models struggle with the syntactic rules of the target language. The analysis of erroneous translations reveals the substantial impact of the interrelationship between involved programming languages and training data on the effectiveness. We show that a general post-processing approach must tolerate inconsistencies and leverage the predictability of LLM answers. Further, we show that iterative translation repair via automated feedback significantly improves translation accuracy. While our combined findings highlight the potential of LLMs to automate code translation, an effective deployment of LLM-based code translation in practice would require models with larger context windows.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-18 in cs.SE).
Categorie arXiv: Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2609.20257 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T04:00:18.366Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).