Metraipolis › archive › dc_70eca407f5a3
RepoNav: From Snippet Retrieval to File-Centered Repository Navigation for Code Agents — arXiv:2609.08355
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.08355
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T12:57:55.406Z
- SHA-256 of the archived text
6c2a77af90ba7832fa83461db44036822ff36130698b93566391067bad26f882- Stable URL of this page
https://metraipolis.com/documento/dc_70eca407f5a3
How to cite
RepoNav: From Snippet Retrieval to File-Centered Repository Navigation for Code Agents — arXiv:2609.08355 — arXiv (en), https://arxiv.org/abs/2609.08355 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_70eca407f5a3 (Metraipolis archive dc_70eca407f5a3, sha256 6c2a77af90ba7832fa83461db44036822ff36130698b93566391067bad26f882)
BibTeX
@misc{dc70eca407f5a3,
title = {RepoNav: From Snippet Retrieval to File-Centered Repository Navigation for Code Agents — arXiv:2609.08355},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.08355},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_70eca407f5a3; sha256 6c2a77af90ba7832fa83461db44036822ff36130698b93566391067bad26f882},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.08355. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Solving repository-level code tasks requires LLM-based agents to use code search tools to navigate large codebases and identify a small set of relevant files and functions. However, current retrieval tools typically return flat lists of isolated code snippets: such lists can surface relevant files, but provide insufficient structure for agents to distinguish the target function from semantically similar alternatives in the same file. We introduce RepoNav, a lightweight post-retrieval interface that reorganizes retrieved snippets into a file-centered navigation scaffold. By presenting compact structural cues and candidate targets, this scaffold guides on-demand file-structure browsing, helping agents compare sibling symbols before selecting a target function. Across diverse models on LocBench, RepoNav improves function-level localization and narrows the file-to-function gap. Controlled ablations demonstrate that these gains come from structured evidence organization rather than simply exposing additional file structure, and the approach also improves performance on a repository-level question-answering benchmark. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL). Categorie arXiv: Software Engineering (cs.SE) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.08355 — 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 2026-10-16T12:30:25.350Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).