Text processing like humans do: Visually attacking and shielding NLP systems
Eger S, Şahin GG, Rücklé A, Lee JU, Schulz C, Mesgar M, Swarnkar K, Simpson E, Gurevych I (2019)
Publication Type: Conference contribution
Publication year: 2019
Publisher: Association for Computational Linguistics (ACL)
Book Volume: 1
Pages Range: 1634-1647
Conference Proceedings Title: NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference
Event location: Minneapolis, MN, USA
ISBN: 9781950737130
Abstract
Visual modifications to text are often used to obfuscate offensive comments in social media (e.g., “!d10t”) or as a writing style (“1337” in “leet speak”), among other scenarios. We consider this as a new type of adversarial attack in NLP, a setting to which humans are very robust, as our experiments with both simple and more difficult visual perturbations demonstrate. We investigate the impact of visual adversarial attacks on current NLP systems on character-, word-, and sentence-level tasks, showing that both neural and non-neural models are, in contrast to humans, extremely sensitive to such attacks, suffering performance decreases of up to 82%. We then explore three shielding methods-visual character embeddings, adversarial training, and rule-based recovery-which substantially improve the robustness of the models. However, the shielding methods still fall behind performances achieved in non-attack scenarios, which demonstrates the difficulty of dealing with visual attacks.
Involved external institutions
Germany (DE)How to cite
APA:
Eger, S., Şahin, G.G., Rücklé, A., Lee, J.U., Schulz, C., Mesgar, M.,... Gurevych, I. (2019). Text processing like humans do: Visually attacking and shielding NLP systems. In NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference (pp. 1634-1647). Minneapolis, MN, USA: Association for Computational Linguistics (ACL).
MLA:
Eger, Steffen, et al. "Text processing like humans do: Visually attacking and shielding NLP systems." Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019, Minneapolis, MN, USA Association for Computational Linguistics (ACL), 2019. 1634-1647.
BibTeX: Download