Metraipolis › archive › dc_98973d2201e5

CrossSWE: Benchmarking Cross-Repository Joint Development for Coding Agents — arXiv:2611.01484

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.01484
License (as archived)
CC0-1.0
Archived
2026-11-03T11:27:18.240Z
SHA-256 of the archived text
25d7dad31be4d317aad28ce85bd33ad93c65ca65b9288af3d23ce75ab16c8b67
Stable URL of this page
https://metraipolis.com/documento/dc_98973d2201e5

How to cite

CrossSWE: Benchmarking Cross-Repository Joint Development for Coding Agents — arXiv:2611.01484 — arXiv (en), https://arxiv.org/abs/2611.01484 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_98973d2201e5 (Metraipolis archive dc_98973d2201e5, sha256 25d7dad31be4d317aad28ce85bd33ad93c65ca65b9288af3d23ce75ab16c8b67)
BibTeX
@misc{dc98973d2201e5,
  title = {CrossSWE: Benchmarking Cross-Repository Joint Development for Coding Agents — arXiv:2611.01484},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.01484},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_98973d2201e5; sha256 25d7dad31be4d317aad28ce85bd33ad93c65ca65b9288af3d23ce75ab16c8b67},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01484. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As coding agents achieve increasingly high success rates on software engineering benchmarks, recent efforts have sought to increase task complexity through longer horizons and more extensive code modifications. However, real-world software development often spans multiple interdependent repositories, introducing additional challenges in dependency reasoning, change propagation, and interface consistency. Rather than further increasing complexity within individual repositories, we introduce CrossSWE, a benchmark for cross-repository joint development comprising 171 real-world instances spanning 165 repositories, 20 technical domains, and 13 programming languages. CrossSWE verifies the necessity of coordinated changes by testing whether partial repository modifications are insufficient to achieve the intended behavior. Our evaluation of five LLMs reveals substantial challenges, with the strongest model, GPT-6.1 Sol, achieving only 58.48% Pass@1. Furthermore, cross-repository inconsistencies emerge as the most prevalent failure mode (44.2%), highlighting the difficulty of maintaining consistency across interdependent repositories.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.SE, nuovi lavori del giorno).
Categorie arXiv: Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2611.01484 — 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 2026-11-03T11:02:52.189Z dalla pagina https://arxiv.org/list/cs.SE/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).