arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2002.00760cs.CLcs.CR
收藏

FastWordBug:一种针对NLP应用程序生成对抗性文本的快速方法

FastWordBug: A Fast Method To Generate Adversarial Text Against NLP Applications

Baidu X-Lab(百度研究院深度学习实验室)

Dou Goodman, Lv Zhonghou, Wang minghua

摘要:

在本文中,我们提出了一种新的算法,FastWordBug,有效地产生小文本扰动的黑盒设置,迫使一个情感分析或文本分类模式作出不正确的预测。通过结合词的词性属性,提出了一种能够快速识别影响文本分类的重要词的评分方法。我们评估FastWordBug在三个真实世界的文本数据集和两个国家的最先进的机器学习模型下黑盒设置。结果表明,我们的方法可以显著降低模型的准确性,同时可以尽可能少地调用模型,具有最高的攻击效率。我们还攻击了两个流行的真实世界的NLP云服务,结果表明,我们的方法也工作。

英文摘要:

In this paper, we present a novel algorithm, FastWordBug, to efficiently generate small text perturbations in a black-box setting that forces a sentiment analysis or text classification mode to make an incorrect prediction. By combining the part of speech attributes of words, we propose a scoring method that can quickly identify important words that affect text classification. We evaluate FastWordBug on three real-world text datasets and two state-of-the-art machine learning models under black-box setting. The results show that our method can significantly reduce the accuracy of the model, and at the same time, we can call the model as little as possible, with the highest attack efficiency. We also attack two popular real-world cloud services of NLP, and the results show that our method works as well.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑