CertiCoder: Towards MISRA-Compliant C Code Generation with LLMs
Proceedings of the ACM on Software Engineering · 30.06.2026 · 3 (FSE) · 547-567
DOI: 10.1145/3797120 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the ACM on Software Engineering, online erschienen am 30.06.2026. Verfasst von Min Gou und 5 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Large language models (LLMs) are increasingly applied to code generation in IDEs, CI pipelines, and automated workflows. Existing evaluations, however, have largely focused on functionality, with comparatively limited attention to compliance with established safety standards. This gap is particularly critical for C, where programmes may compile and pass unit tests yet still violate MISRA C:2012, a widely adopted guideline in safety-critical domains. We present CertiCoder, a post-training framework with rule-aware optimization that transforms tool-verified outcomes into per-rule contrasts and trains models through three stages: rule tuning, cold-start supervised fine-tuning, and rule-aware preference optimization. This design helps models not only distinguish compliant from violating outputs but also associate violations with specific rules. To support reproducible assessment, we construct a Codeforces-derived C benchmark with frozen splits, multi-level decontamination, and metrics that jointly measure MISRA compliance ( S 1 ), functional correctness ( F 1 ), and their conjunction ( J 1 ). On Qwen2.5-Coder backbones (3B–14B), CertiCoder substantially improves compliance from near-zero to measurable J 1 levels and generally preserves functional correctness, outperforming non–rule-aware baselines such as SFT and SafeCoder. To our knowledge, this makes CertiCoder among the first post-training frameworks to explicitly optimize both compliance and correctness, offering a practical step toward more auditable and extensible use of LLMs in safety-critical software systems.
Zitieren
APA
Gou, M., Yao, Z., Ma, H., Zhang, E., Zhou, J., & He, F. (2026). CertiCoder: Towards MISRA-Compliant C Code Generation with LLMs. Proceedings of the ACM on Software Engineering, 3(FSE), 547-567. https://doi.org/10.1145/3797120
Vancouver
Gou M, Yao Z, Ma H, Zhang E, Zhou J, He F. CertiCoder: Towards MISRA-Compliant C Code Generation with LLMs. Proceedings of the ACM on Software Engineering. 2026;3(FSE):547-567. doi:10.1145/3797120
BibTeX
@article{Gou2026,
author = {Gou, Min and Yao, Zhiyu and Ma, Hualong and Zhang, Ende and Zhou, Jian and He, Fei},
title = {CertiCoder: Towards MISRA-Compliant C Code Generation with LLMs},
journal = {Proceedings of the ACM on Software Engineering},
year = {2026},
volume = {3},
number = {FSE},
pages = {547--567},
doi = {10.1145/3797120}
}
EUVIMED