Document Open Access Logo

Balancing Green and Clean Code: Prompting for LLM-Based Refactoring

Authors Feray Gulu-zada , Hina Anwar

PDF

File

Thumbnail PDF
PDF
LIPIcs.ESEM.2026.29.pdf
  • Filesize: 2.64 MB
  • 20 pages

Document Identifiers

Subject Classification

ACM Subject Classification
  • Software and its engineering → Software evolution
Keywords
  • Large language models
  • code refactoring
  • energy efficiency
  • maintainability
  • green software
  • prompt engineering

Metrics

Abstract

Background. LLMs are increasingly used for code refactoring, but their effects on multiple software quality dimensions remain unclear. Green coding targets energy efficiency, while clean coding emphasizes readability and maintainability. These objectives may interact or conflict, making LLM-based refactoring a multi-objective problem. 

Aims. This study investigates whether prompting can guide LLM-based refactoring toward balanced green and clean code outcomes. We examine how different prompting strategies affect energy consumption, runtime performance, and maintainability. 

Method. We conduct a controlled empirical study on 15 human-written Python scripts using four LLMs and five prompting strategies: clean, green, multi-objective, clean-to-green staged, and green-to-clean staged prompting. Generated refactorings are validated for correctness before analysis. Valid outputs are evaluated using CPU energy consumption, runtime, maintainability index, cyclomatic complexity, source lines of code, and Halstead volume. We analyze results using descriptive statistics, statistical tests, correlation analysis, bootstrap confidence intervals, and Pareto-efficiency analysis. 

Results. Of 300 generated refactorings, 267 passed validation. Among valid outputs, 53.18% (142 of 267) reduced energy consumption, but the overall energy effect was small, variable, and not statistically significant across prompting strategies. Energy and runtime changes showed a moderate positive correlation, while maintainability showed weaker relationships with efficiency metrics. Prompting strategies significantly affected maintainability but not energy or runtime. Clean and multi-objective prompting produced the most balanced Pareto outcomes, though these differences were mainly driven by maintainability. 

Conclusions. LLM-based refactoring can sometimes reduce energy consumption without necessarily degrading maintainability, but its effects are inconsistent and depend on model choice, prompt design, and correctness validation. Prompting is useful for steering LLM-based refactoring, but current evidence supports cautious, validated use rather than assuming reliable multi-objective optimization.

Cite As Get BibTex

Feray Gulu-zada and Hina Anwar. Balancing Green and Clean Code: Prompting for LLM-Based Refactoring. In 20th International Symposium on Empirical Software Engineering and Measurement (ESEM 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 394, pp. 29:1-29:20, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026) https://doi.org/10.4230/LIPIcs.ESEM.2026.29

Author Details

Feray Gulu-zada
  • University of Tartu, Estonia
Hina Anwar
  • University of Tartu, Estonia

Acknowledgements

This work was carried out as part of the first author’s thesis research.

Supplementary Materials

References

  1. R. Apsan, V. Stoico, M. Albonico, R. Dhar, K. Vaidhyanathan, and I. Malavolta. Generating energy-efficient code via large-language models – where are we now? In ICSE 2026, 2026. URL: https://doi.org/10.1145/3744916.3764522.
  2. H. Ashraf, S. M. Danish, A. Leivadeas, Y. Otoum, and Z. Sattar. Energy-aware code generation with llms: Benchmarking small vs. large language models for sustainable ai programming. arXiv preprint arXiv:2508.08332, 2025. URL: https://doi.org/10.48550/arXiv.2508.08332.
  3. T. Cappendijk, P. de Reus, and A. Oprescu. An exploration of prompting llms to generate energy-efficient code. In GREENS 2025, pages 31-38, 2025. URL: https://doi.org/10.1109/GREENS66463.2025.00010.
  4. J. Choi, G. An, and S. Yoo. Iterative refactoring of real-world open-source programs with large language models. In Search-Based Software Engineering, pages 49-55. Springer, 2024. URL: https://doi.org/10.1007/978-3-031-64573-0_4.
  5. Tristan Coignion, Clément Quinton, and Romain Rouvoy. When faster isn’t greener: The hidden costs of llm-based code optimization. In 2025 40th IEEE/ACM International Conference on Automated Software Engineering (ASE), pages 1655-1666, 2025. URL: https://doi.org/10.1109/ASE63991.2025.00139.
  6. Jonathan Cordeiro, Shayan Noei, and Ying Zou. An empirical study on the code refactoring capability of large language models. ACM Transactions on Software Engineering and Methodology, 2026. URL: https://doi.org/10.1145/3801158.
  7. V.-A. Cursaru, L. Duits, J. Milligan, D. Ural, B. R. Sanchez, V. Stoico, and I. Malavolta. A controlled experiment on the energy efficiency of the source code generated by code llama. arXiv preprint arXiv:2405.03616, 2024. URL: https://doi.org/10.48550/arXiv.2405.03616.
  8. M. T. Dearing, Y. Tao, X. Wu, Z. Lan, and V. Taylor. Leveraging llms to automate energy-aware refactoring of parallel scientific codes. arXiv preprint arXiv:2505.02184, 2025. URL: https://doi.org/10.48550/arXiv.2505.02184.
  9. K. DePalma, I. Miminoshvili, C. Henselder, K. Moss, and E. A. AlOmar. Exploring chatgpt’s code refactoring capabilities: An empirical study. Expert Systems with Applications, 249:123602, 2024. URL: https://doi.org/10.1016/j.eswa.2024.123602.
  10. M. A. Islam et al. Evaluating the energy-efficiency of the code generated by llms. arXiv preprint arXiv:2505.20324, 2025. URL: https://doi.org/10.48550/arXiv.2505.20324.
  11. S. Garg, R. Zilouchian Moghaddam, and N. Sundaresan. Rapgen: An approach for fixing code inefficiencies in zero-shot. arXiv preprint arXiv:2306.17077, 2025. URL: https://doi.org/10.48550/arXiv.2306.17077.
  12. R. Jonnala, J. Yang, Y. Lee, G. Liang, and Z. Cao. Measuring and improving the efficiency of python code generated by llms using cot prompting and fine-tuning. IEEE Access, 13:1-10, 2025. URL: https://doi.org/10.1109/ACCESS.2025.3585742.
  13. S. H. Kannangara and W. M. J. I. Wijayanake. An empirical evaluation of impact of refactoring on internal and external measures of code quality. International Journal of Software Engineering and Applications, 6(1):1-12, January 2015. Google Scholar
  14. B. Liu, Y. Jiang, Y. Zhang, N. Niu, G. Li, and H. Liu. Exploring the potential of general-purpose llms in automated software refactoring: An empirical study. Automated Software Engineering, 32:26, 2025. URL: https://doi.org/10.1007/s10515-025-00500-0.
  15. Y. Liu, T. Le-Cong, R. Widyasari, C. Tantithamthavorn, L. Li, X.-B. D. Le, and D. Lo. Refining chatgpt-generated code: Characterizing and mitigating code quality issues. ACM, 2023. Google Scholar
  16. Lizard Developers. Lizard: A cyclomatic complexity analyzer. https://pypi.org/project/lizard/, 2026.
  17. Microsoft. Code metrics - maintainability index range and meaning. Microsoft Learn, 2025. URL: https://learn.microsoft.com/en-us/visualstudio/code-quality/code-metrics-maintainability-index-range-and-meaning?view=visualstudio.
  18. H. Nunes, E. Figueiredo, L. Rocha, S. Nadi, F. Ferreira, and G. Esteves. Evaluating the effectiveness of llms in fixing maintainability issues in real-world projects. In IEEE SANER, pages 1-10, 2025. URL: https://doi.org/10.1109/SANER64311.2025.00069.
  19. K. Oueslati, M. Lamothe, and F. Khomh. Refagent: A multi-agent llm-based framework for automatic software refactoring. arXiv preprint arXiv:2511.03153, 2025. URL: https://doi.org/10.48550/arXiv.2511.03153.
  20. G. Pandini, A. Martini, A. N. Videsjorden, and F. A. Fontana. An exploratory study on architectural smell refactoring using large language models. In IEEE International Conference on Software Architecture Companion (ICSA-C), pages 1-10, 2025. URL: https://doi.org/10.1109/ICSA-C65153.2025.00070.
  21. H. Peng, A. Gupte, N. J. Eliopoulos, C. C. Ho, R. Mantri, L. Deng, W. Jiang, and Y.-H. Lu. Large language models for energy-efficient code: Emerging results and future directions. arXiv preprint arXiv:2410.09241, 2024. URL: https://doi.org/10.48550/arXiv.2410.09241.
  22. Y. C. K. Piao, J. C. Paul, L. D. Silva, A. M. Dakhel, M. Hamdaqa, and F. Khomh. Refactoring with llms: Bridging human expertise and machine understanding. arXiv preprint arXiv:2510.03914, 2025. URL: https://doi.org/10.48550/arXiv.2510.03914.
  23. S. Podder, H. Date, and S. Murthy. An empirical study: Leveraging prompt engineering with ai coding assistants to develop energy-efficient code, 2025. Google Scholar
  24. Pylint Developers. Pylint documentation. https://pylint.readthedocs.io/, 2026.
  25. Radon Developers. Radon documentation. https://radon.readthedocs.io/, 2026.
  26. P. Rani, J.-A. Bard, J. Sallou, A. Boll, T. Kehrer, and A. Bacchelli. Can we make code green? understanding trade-offs in llms vs. human code optimizations. arXiv preprint arXiv:2503.20126, 2025. URL: https://doi.org/10.48550/arXiv.2503.20126.
  27. R. Rubei, A. Moussaid, C. Di Sipio, and D. Di Ruscio. Prompt engineering and its implications on the energy consumption of large language models. In IEEE International Conference on Software Analysis, Evolution, and Reengineering (SANER), pages 1-10, 2025. URL: https://ieeexplore.ieee.org/document/11039291.
  28. J. Scheffler, M. Westing, and V. Wohlgemuth. Eneragent: Energy-aware code refinement with llm-based agents. In EnviroInfo, pages 111-118, 2025. Google Scholar
  29. A. Shirafuji, Y. Oda, J. Suzuki, M. Morishita, and Y. Watanobe. Refactoring programs using large language models with few-shot examples. arXiv preprint arXiv:2311.11690, 2023. URL: https://doi.org/10.48550/arXiv.2311.11690.
  30. S. Siddeeq, M. Waseem, Z. Rasheed, M. M. Hasan, J. Rasku, M. Saari, H. Terho, K. Mäkelä, K.-K. Kemell, and P. Abrahamsson. Llm-based multi-agent system for intelligent refactoring of haskell code. arXiv preprint arXiv:2506.19481, 2025. URL: https://doi.org/10.48550/arXiv.2506.19481.
  31. L. Solovyeva, S. Weidmann, and F. Castor. Ai-powered, but power-hungry? energy efficiency of llm-generated code. arXiv preprint arXiv:2502.02412, 2025. URL: https://doi.org/10.48550/arXiv.2502.02412.
  32. J. F. Tuttle, D. Chen, A. Nasrin, N. Soto, and Z. Zong. Can llms generate green code: A comprehensive study through leetcode. In IEEE Green and Sustainable Computing Conference (IGSC), pages 1-10, 2024. URL: https://doi.org/10.1109/IGSC64514.2024.00017.
  33. T. Vartziotis, I. Dellatolas, G. Dasoulas, M. Schmidt, F. Schneider, T. Hoffmann, S. Kotsopoulos, and M. Keckeisen. Learn to code sustainably: An empirical study on llm-based green code generation. arXiv preprint arXiv:2403.03344, 2024. URL: https://doi.org/10.48550/arXiv.2403.03344.
  34. N. Wadhwa, J. Pradhan, A. Sonwane, S. P. Sahu, N. Natarajan, A. Kanade, S. Parthasarathy, and S. Rajamani. Core: Resolving code quality issues using llms. Proceedings of ACM Software Engineering, 1(FSE), 2024. URL: https://doi.org/10.1145/3643762.
  35. S. Waghjale, V. Veerendranath, Z. Z. Wang, and D. Fried. Ecco: Can we improve model-generated code efficiency without sacrificing functional correctness? arXiv preprint arXiv:2407.14044, 2024. URL: https://doi.org/10.48550/arXiv.2407.14044.
  36. B. Zhang, P. Liang, Q. Feng, Y. Fu, and Z. Li. Copilot-in-the-loop: Fixing code smells in copilot-generated python code using copilot. In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering (ASE '24), pages 1-5, 2024. Google Scholar
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact