BBAEG: Towards BERT-based Biomedical Adversarial Example Generation for Text Classification

TL;DR

This paper introduces BBAEG, a black-box attack method for biomedical text classification that combines domain-specific synonym replacement and BERTMLM to generate more effective adversarial examples, improving robustness testing.

Contribution

The paper presents a novel BERT-based adversarial example generation method tailored for biomedical texts, addressing unique challenges in biomedical literature and demonstrating superior attack performance.

Findings

BBAEG outperforms prior methods in attack strength.

Generated adversaries maintain high language fluency and semantic coherence.

Evaluation on biomedical datasets confirms effectiveness.

Abstract

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.