Beyond Language Boundaries: Uncovering Programming Language Families for Code Language Models
Proceedings of the ACM on Software Engineering · 30.06.2026 · 3 (FSE) · 459-478
DOI: 10.1145/3797138 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the ACM on Software Engineering, online erschienen am 30.06.2026. Verfasst von Shangbo Yun und 3 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, Natural Science Foundation of Shanghai.
Abstract (Originalsprache)
The rapid proliferation of diverse programming languages presents both opportunities and challenges for developing multilingual code LLMs. While existing techniques often train code LLMs by simply aggregating multilingual code data, few explore the deeper relationships between programming languages and how such relationships can be utilized to optimize both training and inference. In this work, we investigate two fundamental questions: (1) What are the deep linguistic relationships among programming languages? and (2) How can these relationships be leveraged to improve multilingual code LLMs? We propose an embedding-based framework to uncover the latent families of programming languages. Our approach begins by defining 21 primary linguistic features of programming languages, such as variable definition, control structures, and method declarations, and then employs LLMs to generate feature-aligned code samples across multiple languages. By embedding these semantically parallel code snippets from 19 languages, we construct a similarity matrix and perform hierarchical clustering to uncover inherent language relationships. Our analysis reveals clear hierarchical structures among programming languages. Closely related languages form well-defined clusters (e.g., C, C++, Java, and Swift group together), while Go exhibits as a “lingua franca” with the highest cross-language similarity. Building on the uncovered language families, we propose three strategies to enhance multilingual LLM training: transfer learning across linguistically related languages, linguistic proximity-guided curriculum learning, and centroid-based intermediary code translation. Experiments on four code intelligence tasks demonstrate that our methods significantly improve multilingual LLM performance. This work offers a universal perspective on programming languages and advances more effective strategies for multilingual code LLM training.
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Natural Science Foundation of Shanghai
Zitieren
APA
Yun, S., Gu, X., Huang, J., & Shen, B. (2026). Beyond Language Boundaries: Uncovering Programming Language Families for Code Language Models. Proceedings of the ACM on Software Engineering, 3(FSE), 459-478. https://doi.org/10.1145/3797138
Vancouver
Yun S, Gu X, Huang J, Shen B. Beyond Language Boundaries: Uncovering Programming Language Families for Code Language Models. Proceedings of the ACM on Software Engineering. 2026;3(FSE):459-478. doi:10.1145/3797138
BibTeX
@article{Yun2026,
author = {Yun, Shangbo and Gu, Xiaodong and Huang, Jianghong and Shen, Beijun},
title = {Beyond Language Boundaries: Uncovering Programming Language Families for Code Language Models},
journal = {Proceedings of the ACM on Software Engineering},
year = {2026},
volume = {3},
number = {FSE},
pages = {459--478},
doi = {10.1145/3797138}
}
EUVIMED