BBAEG: Towards BERT-based Biomedical Adversarial Example Generation for Text Classification
This addresses the need for robust predictive models in healthcare analytics, though it is incremental as it adapts existing adversarial generation methods to the biomedical domain.
The paper tackles the vulnerability of deep learning models in biomedical text classification to adversarial examples by proposing BBAEG, a black-box attack algorithm that combines domain-specific synonym replacement, BERT predictions, and other perturbations, demonstrating stronger attacks with better fluency and coherence compared to prior work.
Healthcare predictive analytics aids medical decision-making, diagnosis prediction and drug review analysis. Therefore, prediction accuracy is an important criteria which also necessitates robust predictive language models. However, the models using deep learning have been proven vulnerable towards insignificantly perturbed input instances which are less likely to be misclassified by humans. Recent efforts of generating adversaries using rule-based synonyms and BERT-MLMs have been witnessed in general domain, but the ever increasing biomedical literature poses unique challenges. We propose BBAEG (Biomedical BERT-based Adversarial Example Generation), a black-box attack algorithm for biomedical text classification, leveraging the strengths of both domain-specific synonym replacement for biomedical named entities and BERTMLM predictions, spelling variation and number replacement. Through automatic and human evaluation on two datasets, we demonstrate that BBAEG performs stronger attack with better language fluency, semantic coherence as compared to prior work.