paper-with-me

홈 › Papers

Perturbations in the Wild: Leveraging Human-Written Text Perturbations for Realistic Adversarial Attack and Defense

2022-03-19 · Findings (ACL) 2022 5 · Thai Le, Jooyoung Lee, Kevin Yen, Yifan Hu, Dongwon Lee

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.

📄 PDF Abstract BibTeX arXiv:2203.10346

Code (1)

lethaiq/perturbations-in-the-wild 공식 구현

Tasks

Adversarial Attack

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Residual Connection 설명 없음
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

Perturbations in the Wild: Leveraging Human-Written Text Perturbations for Realistic Adversarial Attack and Defense

2021-11-16 · ACL ARR November 2021 11 · Anonymous

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 Attack

CRYPTEXT: Database and Interactive Toolkit of Human-Written Text Perturbations in the Wild

2023-01-16 · Thai Le, Ye Yiran, Yifan Hu, Dongwon Lee

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

2023-03-18 · Yiran Ye, Thai Le, Dongwon Lee

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 Detection

WildHandBench: A Benchmark for Handwritten Text Understanding that Challenges MLLMs and Humans

2026-08-24 · Jun Zhang, Qiao Zhao, Cheng Cui, Jianying Qu 외 arxiv

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

2024-07-12 · Ayush Singh, Navpreet Singh, Shubham Vatsal

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 …