| Tweet | |
| Kyogo Horikawa, Kosei Horikawa, Yutaro Kashiwa, Hidetake Uwano, and H. Iida, "Do Ai Agents Really Improve Code Readability?," In Proceedings of the 23rd International Conference on Mining Software Repositories (MSR2026), 863–867, July 2026. | |
| ID | 263 |
| 分類 | 国際会議 |
| タグ | agents ai code do improve readability? really |
| 表題 (title) |
Do Ai Agents Really Improve Code Readability? |
| 表題 (英文) |
|
| 著者名 (author) |
Kyogo Horikawa, Kosei Horikawa, Yutaro Kashiwa, Hidetake Uwano, and Hajimu Iida |
| 英文著者名 (author) |
and Hajimu Iida |
| 編者名 (editor) |
|
| 編者名 (英文) |
|
| キー (key) |
and Hajimu Iida |
| 書籍・会議録表題 (booktitle) |
Proceedings of the 23rd International Conference on Mining Software Repositories (MSR2026) |
| 書籍・会議録表題(英文) |
|
| 巻数 (volume) |
|
| 号数 (number) |
|
| ページ範囲 (pages) |
863–867 |
| 組織名 (organization) |
|
| 出版元 (publisher) |
|
| 出版元 (英文) |
|
| 出版社住所 (address) |
|
| 刊行月 (month) |
7 |
| 出版年 (year) |
2026 |
| 採択率 (acceptance) |
|
| URL |
|
| 付加情報 (note) |
|
| 注釈 (annote) |
|
| 内容梗概 (abstract) |
Code readability is fundamental to software quality and maintainability. Poor readability extends development time, increases bug-inducing risks, and contributes to technical debt. With the rapid advancement of Large Language Models, AI agent-based approaches have emerged as a promising paradigm for automated refactoring, capable of decomposing complex tasks through autonomous planning and execution. While prior studies have examined refactoring by AI agents, these analyses cover all forms of refactoring, including performance optimization and structural improvement. As a result, the extent to which AI agent-based refactoring specifically improves code readability remains unclear.
This study investigates the impact of AI agent-based refactoring on code readability. We extracted commits containing readability-related keywords from the AIDev dataset and analyzed changes in readability metrics before and after each commit, covering 403 commits evaluated using multiple quantitative metrics. Our results indicate that AI agents primarily target logic complexity (42.4%) and documentation improvements (24.2%) rather than surface-level aspects like naming conventions or formatting. However, contrary to expectations, readability-focused commits often degraded traditional quality metrics: the Maintainability Index decreased in 56.1% of commits, while Cyclomatic Complexity increased in 42.7%. |
| 論文電子ファイル | 利用できません． |
| BiBTeXエントリ |
@inproceedings{id263,
title = {Do AI Agents Really Improve Code Readability?},
author = {Kyogo Horikawa, Kosei Horikawa, Yutaro Kashiwa, Hidetake Uwano, and Hajimu Iida},
booktitle = {Proceedings of the 23rd International Conference on Mining Software Repositories (MSR2026)},
pages = {863–867},
month = {7},
year = {2026},
}
|
