Bypassing LLM Watermarks with Color-Aware Substitutions
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929285209522176 |
|---|---|
| author | Wu, Qilong Chandrasekaran, Varun |
| author_facet | Wu, Qilong Chandrasekaran, Varun |
| contents | Watermarking approaches are proposed to identify if text being circulated is human or large language model (LLM) generated. The state-of-the-art watermarking strategy of Kirchenbauer et al. (2023a) biases the LLM to generate specific (``green'') tokens. However, determining the robustness of this watermarking method is an open problem. Existing attack methods fail to evade detection for longer text segments. We overcome this limitation, and propose {\em Self Color Testing-based Substitution (SCTS)}, the first ``color-aware'' attack. SCTS obtains color information by strategically prompting the watermarked LLM and comparing output tokens frequencies. It uses this information to determine token colors, and substitutes green tokens with non-green ones. In our experiments, SCTS successfully evades watermark detection using fewer number of edits than related work. Additionally, we show both theoretically and empirically that SCTS can remove the watermark for arbitrarily long watermarked text. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_14719 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Bypassing LLM Watermarks with Color-Aware Substitutions Wu, Qilong Chandrasekaran, Varun Cryptography and Security Computer Vision and Pattern Recognition Machine Learning Watermarking approaches are proposed to identify if text being circulated is human or large language model (LLM) generated. The state-of-the-art watermarking strategy of Kirchenbauer et al. (2023a) biases the LLM to generate specific (``green'') tokens. However, determining the robustness of this watermarking method is an open problem. Existing attack methods fail to evade detection for longer text segments. We overcome this limitation, and propose {\em Self Color Testing-based Substitution (SCTS)}, the first ``color-aware'' attack. SCTS obtains color information by strategically prompting the watermarked LLM and comparing output tokens frequencies. It uses this information to determine token colors, and substitutes green tokens with non-green ones. In our experiments, SCTS successfully evades watermark detection using fewer number of edits than related work. Additionally, we show both theoretically and empirically that SCTS can remove the watermark for arbitrarily long watermarked text. |
| title | Bypassing LLM Watermarks with Color-Aware Substitutions |
| topic | Cryptography and Security Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2403.14719 |