A Strong Baseline for Query Efficient Attacks in a Black Box Setting
This work addresses the problem of high query costs in black-box adversarial attacks for NLP, which is incremental as it builds on prior methods to improve efficiency.
The paper tackles the inefficiency of existing black-box adversarial attacks by proposing a query-efficient strategy that reduces query counts by 75% on average across datasets and models, while maintaining or improving success rates in limited query settings.
Existing black box search methods have achieved high success rate in generating adversarial attacks against NLP models. However, such search methods are inefficient as they do not consider the amount of queries required to generate adversarial attacks. Also, prior attacks do not maintain a consistent search space while comparing different search methods. In this paper, we propose a query efficient attack strategy to generate plausible adversarial examples on text classification and entailment tasks. Our attack jointly leverages attention mechanism and locality sensitive hashing (LSH) to reduce the query count. We demonstrate the efficacy of our approach by comparing our attack with four baselines across three different search spaces. Further, we benchmark our results across the same search space used in prior attacks. In comparison to attacks proposed, on an average, we are able to reduce the query count by 75% across all datasets and target models. We also demonstrate that our attack achieves a higher success rate when compared to prior attacks in a limited query setting.