Perturbations in the Wild: Leveraging Human-Written Text Perturbations for Realistic Adversarial Attack and Defense
We proposes a novel algorithm, ANTHRO, that inductively extracts over 600K human-written text perturbations in the wild and leverages them for realistic adversarial attack. Unlike existing character-based attacks which often deductively hypothesize a set of manipulation strategies, our work is grounded on actual observations from real-world texts. We find that adversarial texts generated by ANTHRO achieve the best trade-off between (1) attack success rate, (2) semantic preservation of the original text, and (3) stealthiness--i.e. indistinguishable from human writings hence harder to be flagged as suspicious. Specifically, our attacks accomplished around 83% and 91% attack success rates on BERT and RoBERTa, respectively. Moreover, it outperformed the TextBugger baseline with an increase of 50% and 40% in terms of semantic preservation and stealthiness when evaluated by both layperson and professional human workers. ANTHRO can further enhance a BERT classifier's performance in understanding different variations of human-written toxic texts via adversarial training when compared to the Perspective API.
Code (1)
Tasks
Adversarial AttackMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Perturbations in the Wild: Leveraging Human-Written Text Perturbations for Realistic Adversarial Attack and Defense
We proposes a novel algorithm, ANTHRO, that inductively extracts over 600K human-written text perturbations in the wild and leverages them for realistic adversarial attack. Unlike existing character-based attacks which …
Adversarial AttackCRYPTEXT: Database and Interactive Toolkit of Human-Written Text Perturbations in the Wild
User-generated textual contents on the Internet are often noisy, erroneous, and not in correct forms in grammar. In fact, some online users choose to express their opinions online through carefully perturbed texts, espec…
NoisyHate: Mining Online Human-Written Perturbations for Realistic Robustness Benchmarking of Content Moderation Models
Online texts with toxic content are a clear threat to the users on social media in particular and society in general. Although many platforms have adopted various measures (e.g., machine learning-based hate-speech detect…
Adversarial AttackBenchmarkingHate Speech DetectionWildHandBench: A Benchmark for Handwritten Text Understanding that Challenges MLLMs and Humans
While the top model on OmniDocBench now reaches 96.34% overall on printed-document parsing, the ability of current models to handle challenging handwritten documents remains largely uncharacterized. Existing benchmarks f…
Robustness of LLMs to Perturbations in Text
Having a clean dataset has been the foundational assumption of most natural language processing (NLP) systems. However, properly written text is rarely found in real-world scenarios and hence, oftentimes invalidates the …