Metraipolis › archive › dc_2e7e0cae4e23

Labelling Bug-Fixing Commits with Local Open-Weight Language Models — arXiv:2609.21616

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21616
License (as archived)
CC0-1.0
Archived
2026-10-10T02:04:10.255Z
SHA-256 of the archived text
94dc2dc8900ce52dd3a06a57acff5698bc36b1f8cd713db84408053418f79e7c
Stable URL of this page
https://metraipolis.com/documento/dc_2e7e0cae4e23

How to cite

Labelling Bug-Fixing Commits with Local Open-Weight Language Models — arXiv:2609.21616 — arXiv (en), https://arxiv.org/abs/2609.21616 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2e7e0cae4e23 (Metraipolis archive dc_2e7e0cae4e23, sha256 94dc2dc8900ce52dd3a06a57acff5698bc36b1f8cd713db84408053418f79e7c)
BibTeX
@misc{dc2e7e0cae4e23,
  title = {Labelling Bug-Fixing Commits with Local Open-Weight Language Models — arXiv:2609.21616},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21616},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2e7e0cae4e23; sha256 94dc2dc8900ce52dd3a06a57acff5698bc36b1f8cd713db84408053418f79e7c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21616. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Defect prediction depends on knowing which commits fix bugs, yet the labels that encode this are produced by routes that each introduce noise. Reused benchmarks carry documented data-quality problems, issue-tracker links are biased and the underlying reports are frequently mistyped, and matching keywords in commit messages is a coarse heuristic. This paper examines whether commits can be labelled as bug fixes from their content alone, using open-weight language models that run locally and therefore keep the process reproducible, inexpensive at corpus scale, usable on proprietary code, and independent of any issue tracker. Against datasets of manually validated and curated bug fixes spanning Java, Python, and JavaScript, we compare a keyword baseline with a set of open-weight models of varying size, prompting each with the commit message and the code diff. On the manually validated corpus the keyword baseline recovers fewer than half of the fixes, whereas the open-weight models recover the large majority and outperform it repository by repository with statistical significance, and larger models do not consistently outperform smaller ones. We further show that evaluation corpora without negative examples cannot support a precision-aware comparison of such classifiers. We release the labelling pipeline together with a labelled, multi-language corpus produced by the recommended configuration, as a reproducible silver-standard resource for building current, project-specific datasets.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-21 in cs.SE).
Categorie arXiv: Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2609.21616 — 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 2026-10-10T01:43:39.834Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).