Newest papers

Documentation Drift in AI-Generated Pull Requests

Will De Michele, Kla Tantithamthavorn · 2026 · arXiv

DOI: 10.48550/arXiv.2610.27381 · open-access PDF

French

Les agents de codage de l’IA contribuent désormais à des répertoires logiciels à l’échelle, mais leurs pratiques de documentation restent non caractérisées. Nous analysons 29,669 demandes de tirage générées par l’IA provenant de 2,585 répertoires dans le groupe de données AIDev, mesurant le co-change de documentation : si des modifications de fichiers de code sont accompagnées d’une mise à jour de fichiers de documentation dans le même PR. Nous constatons que 63,2% des PR d’IA changent de code sans mettre à jour aucune documentation. Cela varie fortement selon le type de tâche: 55,2% pour les fonctionnalités et 77,5% pour le refactoring.

Machine translation (Meta M2M100); the English original follows.

Abstract

AI coding agents now contribute to software repositories at scale, yet their documentation practices remain uncharacterised. We analyse 29,669 AI-generated pull requests from 2,585 repositories in the AIDev dataset, measuring documentation co-change: whether code-file modifications are accompanied by a documentation-file update in the same PR. We find that 63.2% of AI PRs change code without updating any documentation. This varies sharply by task type: 55.2% for features and 77.5% for refactoring. Undocumented code files are followed by a fix to the same file 8.6 to 11.7% more often than docum

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.27381, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange