@inproceedings{siera-etal-2026-team,
    title = "Team Poznan at {S}em{E}val-2026 Task 13: Detecting Machine-Generated Code with Multiple Programming Languages, Generators, and Application Scenarios",
    author = "Siera, Dawid  and
      Kaczmarek, Anatol  and
      Kamzela, Wiktor  and
      Dobosz, Adam  and
      Dutkiewicz, Jakub",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.288/",
    pages = "2275--2280",
    ISBN = "979-8-89176-414-9",
    abstract = "Detecting machine-generated code is crucial for maintaining software security and quality. Traditional approaches often rely on stylistic or statistical features, which are increasingly circumvented by advanced code generation models. This paper introduces a novel approach leveraging Graph Neural Networks (GNNs) to capture the structural characteristics of code, representing it as a program dependency graph. We demonstrate that our GNN-based classifier outperforms both traditional and embedding based methods on benchmark datasets, achieving improved accuracy and robustness in identifying code produced by various generation techniques. This work highlights the potential of GNNs for a more structural understanding of code authorship."
}