Towards Better Correctness and Efficiency in Code Generation
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (36) · 30708-30716
DOI: 10.1609/aaai.v40i36.40327 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yunlong Feng und 4 weiteren Autor:innen.
Abstract (Originalsprache)
While code large language models have demonstrated remarkable progress in code generation, the generated code often exhibits poor runtime efficiency, limiting its practical application in performance-sensitive scenarios. To address this limitation, we propose an efficiency-oriented reinforcement learning framework guided by a novel performance reward. Based on this framework, we take a deeper dive into the code efficiency problem, identifying then proposing methods to overcome key bottlenecks: (1) Dynamic exploration overcomes the static data constraints of offline fine-tuning, enabling the discovery of more efficient code implementations. (2) The error-insensitive reinforcement learning method and high-contrast efficiency signals are crucial for mitigating systematic errors and achieving effective optimization. (3) Online exploration is most effective when starting from a high-correctness baseline, as this allows for efficiency improvements without sacrificing accuracy. With these discoveries, we finally propose a two-stage tuning method, which achieves high and balanced performance across correctness and efficiency. The results of experiments show the effectiveness of the method, which improves code correctness by 10.18% and runtime efficiency by 7.75% on a 7B model, achieving performance comparable to much larger model.
Zitieren
APA
Feng, Y., Xu, Y., Xu, X., Hui, B., & Lin, J. (2026). Towards Better Correctness and Efficiency in Code Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(36), 30708-30716. https://doi.org/10.1609/aaai.v40i36.40327
Vancouver
Feng Y, Xu Y, Xu X, Hui B, Lin J. Towards Better Correctness and Efficiency in Code Generation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(36):30708-30716. doi:10.1609/aaai.v40i36.40327
BibTeX
@article{Feng2026,
author = {Feng, Yunlong and Xu, Yang and Xu, Xiao and Hui, Binyuan and Lin, Junyang},
title = {Towards Better Correctness and Efficiency in Code Generation},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {36},
pages = {30708--30716},
doi = {10.1609/aaai.v40i36.40327}
}
EUVIMED