EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DuCodeMark: Dual-Purpose Code Dataset Watermarking via Style-Aware Watermark-Poison Design

Yuchen Chen, Yuan Xiao, Chunrong Fang, Zhenyu Chen, Baowen Xu

Proceedings of the ACM on Software Engineering · 30.06.2026 · 3 (FSE) · 1059-1081

DOI: 10.1145/3797076 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the ACM on Software Engineering, online erschienen am 30.06.2026. Verfasst von Yuchen Chen und 4 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The proliferation of large language models for code (CodeLMs) and open-source contributions has heightened concerns over unauthorized use of source code datasets. While watermarking provides a viable protection mechanism by embedding ownership signals, existing methods rely on detectable trigger-target patterns and are limited to source-code tasks, overlooking other scenarios such as decompilation tasks. In this paper, we propose DuCodeMark, a stealthy and robust dual-purpose watermarking method for code datasets that generalizes across both source-code tasks and decompilation tasks. DuCodeMark parses each code sample into an abstract syntax tree (AST), applies language-specific style transformations to construct stealthy trigger-target pairs, and injects repressible poisoned features into a subset of return-typed samples to enhance robustness against watermark removal or evasion. These features remain inactive during normal training but are activated upon watermark removal, degrading model performance. For verification, DuCodeMark employs a black-box method based on the independent-samples t -test. We conduct a comprehensive evaluation of DuCodeMark across 72 settings spanning two code tasks, two programming languages, three CodeLMs, and six decoding temperatures. The results demonstrate that it consistently achieves strong verifiability ( p < 0.05), high stealthiness (suspicion rate ≤ 0.36), robustness against both watermark and poisoning attacks (recall ≤ 0.57), and a substantial drop in model performance upon watermark removal (Pass@1 drops by 28.6

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Chen, Y., Xiao, Y., Fang, C., Chen, Z., & Xu, B. (2026). DuCodeMark: Dual-Purpose Code Dataset Watermarking via Style-Aware Watermark-Poison Design. Proceedings of the ACM on Software Engineering, 3(FSE), 1059-1081. https://doi.org/10.1145/3797076

Vancouver

Chen Y, Xiao Y, Fang C, Chen Z, Xu B. DuCodeMark: Dual-Purpose Code Dataset Watermarking via Style-Aware Watermark-Poison Design. Proceedings of the ACM on Software Engineering. 2026;3(FSE):1059-1081. doi:10.1145/3797076

BibTeX
@article{Chen2026,
  author = {Chen, Yuchen and Xiao, Yuan and Fang, Chunrong and Chen, Zhenyu and Xu, Baowen},
  title = {DuCodeMark: Dual-Purpose Code Dataset Watermarking via Style-Aware Watermark-Poison Design},
  journal = {Proceedings of the ACM on Software Engineering},
  year = {2026},
  volume = {3},
  number = {FSE},
  pages = {1059--1081},
  doi = {10.1145/3797076}
}
Kostenlos starten →