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 2310.09652cs.CR
收藏

BufferSearch:以更低查询次数生成黑盒对抗文本

BufferSearch: Generating Black-Box Adversarial Texts With Lower Queries

Wenjie Lv, Zhen Wang, Yitao Zheng, Zhehua Zhong, Qi Xuan, Tianyi Chen

AI总结:

本文提出查询高效的 BufferSearch 方法,利用历史信息与统计检验减少黑盒 NLP 对抗攻击中的模型查询次数,在保持攻击性能的同时显著降低查询量,并建立了查询效率研究的基准。

摘要:

机器学习安全近来已成为自然语言处理(NLP)领域的一个突出话题。现有的黑盒对抗攻击因模型查询复杂度极高而备受困扰,导致容易被反攻击监控器捕获。与此同时,如何消除冗余的模型查询却鲜有探索。在本文中,我们提出了一种查询高效的方法 BufferSearch,以最少的查询请求次数有效攻击通用的智能 NLP 系统。总的来说,BufferSearch 利用历史信息并进行统计检验,以避免频繁引发模型查询。在数值上,我们在多种基准文本分类实验中展示了 BufferSearch 的有效性,在取得具有竞争力的攻击性能的同时,显著减少了查询数量。此外,在受限的查询预算内,BufferSearch 的性能比竞争对手高出数倍。我们的工作为未来 NLP 对抗攻击中查询效率的研究建立了强有力的基准。

英文摘要:

Machine learning security has recently become a prominent topic in the natural language processing (NLP) area. The existing black-box adversarial attack suffers prohibitively from the high model querying complexity, resulting in easily being captured by anti-attack monitors. Meanwhile, how to eliminate redundant model queries is rarely explored. In this paper, we propose a query-efficient approach BufferSearch to effectively attack general intelligent NLP systems with the minimal number of querying requests. In general, BufferSearch makes use of historical information and conducts statistical test to avoid incurring model queries frequently. Numerically, we demonstrate the effectiveness of BufferSearch on various benchmark text-classification experiments by achieving the competitive attacking performance but with a significant reduction of query quantity. Furthermore, BufferSearch performs multiple times better than competitors within restricted query budget. Our work establishes a strong benchmark for the future study of query-efficiency in NLP adversarial attacks.

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。

↑