Newest papers

Code Comprehensor: Hierarchical Documentation and Cross-Modal Retrieval for Large Repositories

Sujoy Roychowdhury, Joseph H. R. Isaac, Jiangning Gao, Martin Klitte, Diarmuid Corcoran · 2026 · arXiv

DOI: 10.48550/arXiv.2610.30016 · open-access PDF

German

Das Verständnis großer, komplexer und sich schnell entwickelnder Code-Basen bleibt eine zentrale Herausforderung für die Software-Engineering im Maßstab, die Onboarding, Wartung, Refactoring und Impact-Analyse behindert. Die Bereitstellung dieser Informationen an ein Large-Language Model (LLM) ist ein weiteres Hindernis, da der LLM nicht nur die erforderliche Code-Methode verstehen muss, sondern auch die damit verbundenen Methoden im großen Schema des Repositoriums. Wir schlagen Code Comprehensor vor, ein Hybrid-Recovery-Reasoning-Framework, das (i) ein Code-Dependency-Callgraph, (ii) CodeScribe, das Code-aware LLMs verwendet, um hierarchische Dokumentation Acro zu generieren

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

Abstract

Understanding large, complex, and rapidly evolving code bases remains a core challenge for software engineering at scale, hindering onboarding, maintenance, refactoring, and impact analysis. Providing this information to a Large-Language Model (LLM) is another hurdle, as the LLM must understand not only the required code method but also its associated methods in the grand scheme of the repository. We propose Code Comprehensor, a hybrid retrieval-reasoning framework combining (i) a code dependency callgraph, (ii) CodeScribe, which uses code-aware LLMs to generate hierarchical documentation acro

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=2610.30016, 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