@inproceedings{kim-etal-2026-codestruct,
    title = "{CODESTRUCT}: Code Agents over Structured Action Spaces",
    author = "Kim, Myeongsoo  and
      Hsu, Chao-Chun  and
      Wang, Dingmin  and
      Garg, Shweta  and
      Kumar, Varun  and
      Ramanathan, Murali Krishna",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.607/",
    pages = "13290--13306",
    ISBN = "979-8-89176-390-6",
    abstract = "LLM-based code agents treat repositories as unstructured text, applying edits through brittle string matching that frequently fails due to formatting drift or ambiguous patterns. We propose reframing the codebase as a structured action space where agents operate on named AST entities rather than text spans. Our framework, CodeStruct, provides readCode for retrieving complete syntactic units and editCode for applying syntax-validated transformations to semantic program elements. Evaluated on SWE-Bench Verified across six LLMs, CodeStruct improves Pass@1 accuracy by 1.2-5.0{\%} while reducing token consumption by 12-38{\%} for most models. Models that frequently fail to produce valid patches under text-based interfaces benefit most: GPT-5-nano improves by 20.8{\%} as empty-patch failures drop from 46.6{\%} to 7.2{\%}. On CodeAssistBench, we observe consistent accuracy gains (+0.8-4.4{\%}) with cost reductions up to 33{\%}. Our results show that structure-aware interfaces offer a more reliable foundation for code agents."
}