• Computer Science > Artificial Intelligence [Submitted on 30 Jan 2026] Title:AST-PAC: AST-guided Membership Inference for Code View PDF HTML (experimental)Abstract:Code Large Language Models are frequently trained on massive datasets containing restrictively licensed source code. • This creates urgent data governance and copyright challenges. • Membership Inference Attacks (MIAs) can serve as an auditing mechanism to detect unauthorized data usage in models. • While attacks like the Loss Attack provide a baseline, more involved methods like Polarized Augment Calibration (PAC) remain underexplored in the code domain. • This paper presents an exploratory study evaluating these methods on 3B–7B parameter code models. • We find that while PAC generally outperforms the Loss baseline, its effectiveness relies on augmentation strategies that disregard the rigid syntax of code, leading to performance degradation on larger, complex files.

Article Summaries:

  • Computer Science > Artificial Intelligence [Submitted on 30 Jan 2026] Title:AST-PAC: AST-guided Membership Inference for Code View PDF HTML (experimental)Abstract:Code Large Language Models are frequently trained on massive datasets containing restrictively licensed source code. This creates urgent data governance and copyright challenges. Membership Inference Attacks (MIAs) can serve as an auditing mechanism to detect unauthorized data usage in models. While attacks like the Loss Attack provide a baseline, more involved methods like Polarized Augment Calibration (PAC) remain underexplored in

Sources: