让AI生成文本的水印可按权限查看，保护隐私又不失精准。
Selective Disclosure Watermarking for Large Language Models
- 分层词表路由技术，将水印信息分级存储
- 验证时仅暴露指定部分水印，整体信息不泄露
- 保持文本质量，检测准确率高且响应快
水印技术可在大语言模型生成的文本中嵌入不可察觉且可验证的信号。现有方法包括零比特方案（区分机器与人工文本）和多比特方案（嵌入元数据）。但当前多比特水印无法实现选择性披露：验证任意部分需暴露全部消息，导致信息过度泄露，引发隐私问题。本文提出分层词表路由（HeRo）框架，实现嵌入元数据的选择性披露。该方法递归划分词表，将水印信息分布于多级层次结构中，不同验证者仅能解码其权限范围内的数据。实验表明，该方案保持底层采样过程无偏，文本质量不受影响，同时支持细粒度访问控制，检测准确率高、延迟低。代码已开源：https://github.com/xuyangc03/hero-watermark。
原文摘要 · Abstract (English)
Watermarking methods embed imperceptible and verifiable signals into text generated by large language models (LLMs). Existing approaches include zero-bit schemes for distinguishing synthetic text from human writing and multi-bit schemes for embedding metadata. However, current multi-bit watermarking methods do not allow selective disclosure: verifying any part of the watermark requires revealing the entire embedded message. This lack of control leads to unnecessary information exposure and raises privacy concerns. We propose Hierarchical Vocabulary Routing (HeRo), a watermarking framework that enables selective disclosure of embedded metadata. The method recursively partitions the vocabulary and distributes watermark information across hierarchical layers, so that different verifiers can decode only the portions of the payload corresponding to their access level. We show that the proposed scheme preserves the unbiasedness of the underlying sampling process and thus maintains text quality. Experiments demonstrate that our framework supports fine-grained access control while achieving high detection accuracy and low latency. Code is available at https://github.com/xuyangc03/hero-watermark.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。