Jinfeng Li (Zhejiang University), Shouling Ji (Zhejiang University), Tianyu Du (Zhejiang University), Bo Li (University of California, Berkeley), Ting Wang (Lehigh University)

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 user sentiment analysis and toxic content detection. In this paper, we show that DLTU is inherently vulnerable to adversarial text attacks, in which maliciously crafted text triggers target DLTU systems and services to misbehave. Specifically, we present TextBugger, a general attack framework for generating adversarial text. In contrast of prior work, 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.

View More Papers

Analyzing Semantic Correctness with Symbolic Execution: A Case Study...

Sze Yiu Chau (Purdue University), Moosa Yahyazadeh (The University of Iowa), Omar Chowdhury (The University of Iowa), Aniket Kate (Purdue University), Ninghui Li (Purdue University)

Read More

A Systematic Framework to Generate Invariants for Anomaly Detection...

Cheng Feng (Imperial College London & Siemens Corporate Technology), Venkata Reddy Palleti (Singapore University of Technology and Design), Aditya Mathur (Singapore University of Technology and Design), Deeph Chana (Imperial College London)

Read More

How Bad Can It Git? Characterizing Secret Leakage in...

Michael Meli (North Carolina State University), Matthew R. McNiece (Cisco Systems and North Carolina State University), Bradley Reaves (North Carolina State University)

Read More

Mind Your Own Business: A Longitudinal Study of Threats...

Platon Kotzias (IMDEA Software Institute, Universidad Politécnica de Madrid), Leyla Bilge (Symantec Research Labs), Pierre-Antoine Vervier (Symantec Research Labs), Juan Caballero (IMDEA Software Institute)

Read More