EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CodeS : Natural Language to Code Repository via Multi-Layer Sketch

Daoguang Zan, Ailun Yu, Wei Liu, Dong Chen, Bo Shen, Yafen Yao, Wei Li, Xiaolin Chen, Yongshun Gong, Bei Guan, Zhiguang Yang, Yongji Wang und 2 weitere

ACM Transactions on Software Engineering and Methodology · 31.07.2026 · 35 (7) · 1-26

DOI: 10.1145/3768577 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Software Engineering and Methodology, online erschienen am 17.06.2026. Verfasst von Daoguang Zan und 13 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, “Unified Representation and Knowledge Graph Construction for Science Popularization Resources.”.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The impressive performance of Large Language Models (LLMs) on code-related tasks has shown the potential of fully automated software development. In light of this, we introduce a new software engineering task, namely Natural Language to Code Repository (NL2Repo). This task aims to generate an entire code repository from its natural language requirements. To address this task, we propose a simple yet effective framework CodeS , which decomposes NL2Repo into multiple sub-tasks by a multi-layer sketch. Specifically, CodeS includes three modules: RepoSketcher, FileSketcher, and SketchFiller. RepoSketcher first generates a repository’s directory structure for given requirements; FileSketcher then generates a file sketch for each file in the generated structure; SketchFiller finally fills in the details for each function in the generated file sketch. To rigorously assess CodeS on the NL2Repo task, we carry out evaluations through both automated benchmarking and manual feedback analysis. For benchmark-based evaluation, we craft a repository-oriented benchmark, SketchEval, and design an evaluation metric, SketchBLEU. For feedback-based evaluation, we develop a VSCode plugin for CodeS and engage 30 participants in conducting empirical studies. Extensive experiments prove the effectiveness and practicality of CodeS on the NL2Repo task.

Förderung

  • National Key Research and Development Program of China
  • “Unified Representation and Knowledge Graph Construction for Science Popularization Resources.”

Zitieren

APA

Zan, D., Yu, A., Liu, W., Chen, D., Shen, B., Yao, Y., Li, W., Chen, X., Gong, Y., Guan, B., Yang, Z., Wang, Y., Cui, L., & Wang, Q. (2026). CodeS : Natural Language to Code Repository via Multi-Layer Sketch. ACM Transactions on Software Engineering and Methodology, 35(7), 1-26. https://doi.org/10.1145/3768577

Vancouver

Zan D, Yu A, Liu W, Chen D, Shen B, Yao Y, et al. CodeS : Natural Language to Code Repository via Multi-Layer Sketch. ACM Transactions on Software Engineering and Methodology. 2026;35(7):1-26. doi:10.1145/3768577

BibTeX
@article{Zan2026,
  author = {Zan, Daoguang and Yu, Ailun and Liu, Wei and Chen, Dong and Shen, Bo and Yao, Yafen and Li, Wei and Chen, Xiaolin and Gong, Yongshun and Guan, Bei and Yang, Zhiguang and Wang, Yongji and Cui, Lizhen and Wang, Qianxiang},
  title = {CodeS : Natural Language to Code Repository via Multi-Layer Sketch},
  journal = {ACM Transactions on Software Engineering and Methodology},
  year = {2026},
  volume = {35},
  number = {7},
  pages = {1--26},
  doi = {10.1145/3768577}
}
Kostenlos starten →