2018/12/13 by Jinfeng Li, Shouling Ji, Tianyu Du +2 · 326 citations
Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Adversarial machine learning #Adversarial system #Hate Speech and Cyberbullying Detection #Key (lock) #Semantics (computer science) #Sentiment analysis #Set (abstract data type) #cs.CL #cs.CR #cs.LG
paper · pdf · doi:10.14722/ndss.2019.23138
To appear in NDSS 2019
arxiv created 2018/12/13 · arxiv updated 2018/12/14 · openalex created_date 2018/12/22 · openalex publication_date 2019/01/01 · openalex updated_date 2026/08/05
Deep Learning-based Text Understanding (DLTU) is the backbone technique behind various applications, including question answering, machine translation, and text classification. Despite its tremendous popularity, the security vulnerabilities of DLTU are still largely unknown, which is highly concerning given its increasing use in security-sensitive applications such as sentiment analysis and toxic content detection. In this paper, we show that DLTU is inherently vulnerable to adversarial text attacks, in which maliciously crafted texts trigger target DLTU systems and services to misbehave. Specifically, we present TEXTBUGGER, a general attack framework for generating adversarial texts. In contrast to prior works, TEXTBUGGER differs in significant ways: (i) effective -it outperforms state-of-the-art attacks in terms of attack success rate; (ii) evasive -it preserves the utility of benign text, with 94.9% of the adversarial text correctly recognized by human readers; and (iii) efficient -it generates adversarial text with computational complexity sub-linear to the text length. We empirically evaluate TEXTBUGGER on a set of real-world DLTU systems and services used for sentiment analysis and toxic content detection, demonstrating its effectiveness, evasiveness, and efficiency. For instance, TEXTBUGGER achieves 100% success rate on the IMDB dataset based on Amazon AWS Comprehend within 4.61 seconds and preserves 97% semantic similarity. We further discuss possible defense mechanisms to mitigate such attack and the adversary's potential countermeasures, which leads to promising directions for further research.