arXiv cs.CY· Alexander Nemecek, Yuqiao Xu, Joseph Falkenburg, Erman Ayday·· 11 小时前AI 评分31
无需 Tokenizer 仍有差距：同形字与不可见字符水印的跨语言公平性研究
No Tokenizer, Still a Gap: Cross-Lingual Fairness of Homoglyph and Invisible-Character Watermarking
AI 导读
研究在 24 种文字体系的 40 种语言变体上测量同形字与不可见字符水印，发现隐藏容量相差 3.3 倍，其中 96% 由文字体系解释。拉丁文字内部资源水平无显著差异（ρ=+0.095，p=0.78），跨文字体系则随资源水平上升（ρ=+0.54）。普通复制粘贴会剥离不可见字符并拉大差距：检测器约需 2 个英文或波斯文码点，而老挝文需 93 个。
正文 · 原文
Abstract:Homoglyph and invisible-character watermarks are edit-based watermarks. Instead of changing how a model generates text, they hide a signal in the text after it is written. They swap some characters for homoglyphs, which look the same, or very similar, but have different Unicode code points, or they insert characters that do not display, all without access to the model that wrote the text. Generation-time watermarks work less well for lower-resourced languages due to how tokenizers split their text. Because they work on individual characters, these watermarks never touch a tokenizer, so we ask whether they escape this gap. Across 40 language varieties in 24 scripts, we measure capacity (the hidden bits each character can carry), how much of it survives routine handling and targeted removal, and how much text a keyed detector needs. Capacity varies 3.3-fold and script accounts for 96\% of that variation. Within Latin script, resource level makes no detectable difference ($\rho=+0.095$, $p=0.78$), but across scripts capacity rises with it ($\rho=+0.54$, 95\% CI $[+0.14,+0.83]$), because lower-resourced languages tend to use scripts with few homoglyphs. Ordinary copy-paste, which strips invisible characters, widens the gap: a detector then needs about 2 code points of English or Persian but 93 of Lao. Removing the tokenizer does not remove the inequity, but instead only changes its source.
| Comments: | 19 pages |
| Subjects: | Computation and Language (cs.CL); Cryptography and Security (cs.CR); Computers and Society (cs.CY) |
| Cite as: | arXiv:2610.36432 [cs.CL] |
| (or arXiv:2610.36432v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36432 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Alexander Nemecek [view email]
[v1]
Tue, 27 Oct 2026 00:27:01 UTC (169 KB)
来源：arXiv cs.CY · arxiv.org