HyperparameterOmens at SemEval-2026 Task 13: Various approaches to detecting machine- generated code
Year: 2026
Type: conference-paper
Abstract: We present our systems for SemEval-2026 Task 13 (Orel et al., 2026) (Detecting Machine-Generated Code), built on the Droid resource suite and benchmark setting (Orel et al., 2025).For Subtask A (binary classification of humanwritten vs. machine-generated code), lexical baselines such as TF-IDF and character ngrams failed to generalize from the LeetCode training distribution to the production-code evaluation split.After correcting pipeline errors that obscured true performance and selecting stable AST features under domain shift, our final system uses 5 uncorrelated features and achieves 0.57 macro F1 on the public test set.For Subtask C (4-way authorship classification of human, AI, hybrid, and adversarial), lexical baselines collapsed under vocabulary shift (55% F1 drop), while deep semantic models achieved near-zero degradation.A per-class weighted ensemble achieved 0.85 macro F1 on the 1K public sample, with a final score of 0.57 on the full test set.
Authors: Dmitry Sukhotin, How Yu
Institution: University of Tübingen
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166808083