Protecting intellectual property (IP) of text such as articles and code is increasingly important, especially as sophisticated attacks become possible, such as paraphrasing by large language models (LLMs) or even unauthorized training of LLMs on copyrighted text to infringe such IP. However, existing text watermarking methods are not robust enough against such attacks nor scalable to millions of users for practical implementation. In this paper, we propose Waterfall, the first training-free framework for robust and scalable text watermarking applicable across multiple text types (e.g., articles, code) and languages supportable by LLMs, for general text and LLM data provenance. Waterfall comprises several key innovations, such as being the first to use LLM as paraphrasers for watermarking along with a novel combination of techniques that are surprisingly effective in achieving robust verifiability and scalability. We empirically demonstrate that Waterfall achieves significantly better scalability, robust verifiability, and computational efficiency compared to SOTA article-text watermarking methods, and also showed how it could be directly applied to the watermarking of code.

翻译：保护文章和代码等文本的知识产权（IP）日益重要，尤其是在大型语言模型（LLMs）进行释义攻击、甚至利用受版权保护的文本未经授权训练LLMs以侵犯此类IP等复杂攻击成为可能的情况下。然而，现有的文本水印方法对此类攻击的鲁棒性不足，也难以扩展到数百万用户以实现实际部署。本文提出Waterfall，这是首个无需训练的鲁棒且可扩展的文本水印框架，适用于多种文本类型（例如文章、代码）以及LLMs支持的语言，旨在实现通用文本和LLM数据的来源追溯。Waterfall包含多项关键创新，例如首次将LLM用作水印处理的释义器，并结合了一系列新颖技术，这些技术在实现鲁棒可验证性和可扩展性方面表现出惊人的有效性。我们通过实验证明，与最先进的文章文本水印方法相比，Waterfall在可扩展性、鲁棒可验证性和计算效率方面均显著更优，并展示了其如何直接应用于代码水印。