Style Attack Disguise: When Fonts Become a Camouflage for Adversarial Intent
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914108168732672 |
|---|---|
| author | Zhang, Yangshijie Wang, Xinda Liu, Jialin Wang, Wenqiang Ma, Zhicong Jia, Xingxing |
| author_facet | Zhang, Yangshijie Wang, Xinda Liu, Jialin Wang, Wenqiang Ma, Zhicong Jia, Xingxing |
| contents | With social media growth, users employ stylistic fonts and font-like emoji to express individuality, creating visually appealing text that remains human-readable. However, these fonts introduce hidden vulnerabilities in NLP models: while humans easily read stylistic text, models process these characters as distinct tokens, causing interference. We identify this human-model perception gap and propose a style-based attack, Style Attack Disguise (SAD). We design two sizes: light for query efficiency and strong for superior attack performance. Experiments on sentiment classification and machine translation across traditional models, LLMs, and commercial services demonstrate SAD's strong attack performance. We also show SAD's potential threats to multimodal tasks including text-to-image and text-to-speech generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_19641 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Style Attack Disguise: When Fonts Become a Camouflage for Adversarial Intent Zhang, Yangshijie Wang, Xinda Liu, Jialin Wang, Wenqiang Ma, Zhicong Jia, Xingxing Computation and Language Artificial Intelligence With social media growth, users employ stylistic fonts and font-like emoji to express individuality, creating visually appealing text that remains human-readable. However, these fonts introduce hidden vulnerabilities in NLP models: while humans easily read stylistic text, models process these characters as distinct tokens, causing interference. We identify this human-model perception gap and propose a style-based attack, Style Attack Disguise (SAD). We design two sizes: light for query efficiency and strong for superior attack performance. Experiments on sentiment classification and machine translation across traditional models, LLMs, and commercial services demonstrate SAD's strong attack performance. We also show SAD's potential threats to multimodal tasks including text-to-image and text-to-speech generation. |
| title | Style Attack Disguise: When Fonts Become a Camouflage for Adversarial Intent |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.19641 |