Abstract: Semantics-preserving program transformations (SPTs) are widely used to generate adversarial example attacks against deep learning-based models for code analysis tasks. This work studies 34 ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results