Abstract
Deep learning underpins most of the currently advanced natural language processing (NLP) tasks such as textual classification, neural machine translation (NMT), abstractive summarization and question-answering (QA). However, the robustness of the models, particularly QA models, against adversarial attacks is a critical concern that remains insufficiently explored. This paper introduces QA-Attack (Question Answering Attack), a novel word-level adversarial strategy that fools QA models. Our attention-based attack exploits the customized attention mechanism and deletion ranking strategy to identify and target specific words within contextual passages. It creates deceptive inputs by carefully choosing and substituting synonyms, preserving grammatical integrity while misleading the model to produce incorrect responses. Our approach demonstrates versatility across various question types, particularly when dealing with extensive long textual inputs. Extensive experiments on multiple benchmark datasets demonstrate that QA-Attack successfully deceives baseline QA models and surpasses existing adversarial techniques regarding success rate, semantics changes, BLEU score, fluency and grammar error rate.
Abstract (translated)
深度学习支撑了当前大多数先进的自然语言处理（NLP）任务，如文本分类、神经机器翻译（NMT）、抽象总结和问答（QA）。然而，模型的鲁棒性，尤其是QA模型对抗攻击的鲁棒性，仍然是一个尚未充分研究的重要问题。本文介绍了QA-Attack（问答攻击），一种新颖的词级对抗策略，用于欺骗QA模型。我们的基于注意力的攻击方法利用了定制的注意力机制和删除排名策略来识别并针对上下文段落中的特定单词。通过精心选择和替换同义词，它创建了误导性的输入，在保持语法完整性的同时引导模型生成错误的回答。我们的方法在处理各种问题类型时展现了灵活性，特别是在应对大量长文本输入的情况下尤为突出。在多个基准数据集上的广泛实验表明，QA-Attack成功欺骗了基线的QA模型，并且在成功率、语义变化度、BLEU分数、流畅性和语法错误率方面超过了现有的对抗技术。
URL
https://arxiv.org/abs/2411.08248
