arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.13484cs.LGcs.AIcs.CLcs.CR
收藏

利用目标信息的改进高效文本对抗攻击

Improved and Efficient Text Adversarial Attacks using Target Information

Monash University(莫纳什大学)

Mahmoud Hossam, Trung Le, He Zhao, Viet Huynh, Dinh Phung

摘要:

最近,在黑盒环境中研究自然语言模型上的对抗性示例的兴趣越来越大。这些方法通过干扰某些重要的词直到分类器标签被改变来攻击自然语言分类器。为了找到这些重要的词,这些方法通过对每个输入句子逐词查询目标模型来对所有词进行重要性排名,导致查询效率很高。引入了一种新的有趣的方法,通过可解释的学习来学习单词排名而不是以前的昂贵搜索来解决这个问题。使用这种方法的主要优点是,它实现了与最先进的方法相当的攻击率,但速度更快,查询更少,其中需要更少的查询以避免对攻击代理的怀疑。尽管如此,这种方法牺牲了有用的信息,可以利用从目标分类器的查询效率的缘故。在本文中,我们研究了利用目标模型输出和数据对攻击率和平均查询次数的影响,我们表明这两个方面都可以得到改善,额外查询的开销有限。

英文摘要:

There has been recently a growing interest in studying adversarial examples on natural language models in the black-box setting. These methods attack natural language classifiers by perturbing certain important words until the classifier label is changed. In order to find these important words, these methods rank all words by importance by querying the target model word by word for each input sentence, resulting in high query inefficiency. A new interesting approach was introduced that addresses this problem through interpretable learning to learn the word ranking instead of previous expensive search. The main advantage of using this approach is that it achieves comparable attack rates to the state-of-the-art methods, yet faster and with fewer queries, where fewer queries are desirable to avoid suspicion towards the attacking agent. Nonetheless, this approach sacrificed the useful information that could be leveraged from the target classifier for that sake of query efficiency. In this paper we study the effect of leveraging the target model outputs and data on both attack rates and average number of queries, and we show that both can be improved, with a limited overhead of additional queries.

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。

↑