Model Robustness with Text Classification: Semantic-preserving adversarial attacks
This work addresses the need for robust text classification models against adversarial attacks, but it appears incremental as it builds on existing adversarial attack methods.
The authors tackled the problem of assessing model robustness in text classification by proposing algorithms for semantic-preserving adversarial attacks, which cause significant flips in white-box settings and reverse decisions in black-box transformer-based architectures.
We propose algorithms to create adversarial attacks to assess model robustness in text classification problems. They can be used to create white box attacks and black box attacks while at the same time preserving the semantics and syntax of the original text. The attacks cause significant number of flips in white-box setting and same rule based can be used in black-box setting. In a black-box setting, the attacks created are able to reverse decisions of transformer based architectures.