@inproceedings{eger-benz-2020-hero,
    title = "From Hero to Z{\'e}roe: A Benchmark of Low-Level Adversarial Attacks",
    author = "Eger, Steffen  and
      Benz, Yannik",
    editor = "Wong, Kam-Fai  and
      Knight, Kevin  and
      Wu, Hua",
    booktitle = "Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing",
    month = dec,
    year = "2020",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/add-emnlp-2024-awards/2020.aacl-main.79/",
    doi = "10.18653/v1/2020.aacl-main.79",
    pages = "786--803",
    abstract = "Adversarial attacks are label-preserving modifications to inputs of machine learning classifiers designed to fool machines but not humans. Natural Language Processing (NLP) has mostly focused on high-level attack scenarios such as paraphrasing input texts. We argue that these are less realistic in typical application scenarios such as in social media, and instead focus on low-level attacks on the character-level. Guided by human cognitive abilities and human robustness, we propose the first large-scale catalogue and benchmark of low-level adversarial attacks, which we dub Z{\'e}roe, encompassing nine different attack modes including visual and phonetic adversaries. We show that RoBERTa, NLP`s current workhorse, fails on our attacks. Our dataset provides a benchmark for testing robustness of future more human-like NLP models."
}