|国家预印本平台
首页|CredID: Credible Multi-Bit Watermark for Large Language Models Identification

CredID: Credible Multi-Bit Watermark for Large Language Models Identification

CredID: Credible Multi-Bit Watermark for Large Language Models Identification

来源:Arxiv_logoArxiv
英文摘要

Large Language Models (LLMs) are widely used in complex natural language processing tasks but raise privacy and security concerns due to the lack of identity recognition. This paper proposes a multi-party credible watermarking framework (CredID) involving a trusted third party (TTP) and multiple LLM vendors to address these issues. In the watermark embedding stage, vendors request a seed from the TTP to generate watermarked text without sending the user's prompt. In the extraction stage, the TTP coordinates each vendor to extract and verify the watermark from the text. This provides a credible watermarking scheme while preserving vendor privacy. Furthermore, current watermarking algorithms struggle with text quality, information capacity, and robustness, making it challenging to meet the diverse identification needs of LLMs. Thus, we propose a novel multi-bit watermarking algorithm and an open-source toolkit to facilitate research. Experiments show our CredID enhances watermark credibility and efficiency without compromising text quality. Additionally, we successfully utilized this framework to achieve highly accurate identification among multiple LLM vendors.

Shanzhe Lei、Ping Yi、Xuhong Wang、Haoyu Jiang、Yilun Lin

计算技术、计算机技术

Shanzhe Lei,Ping Yi,Xuhong Wang,Haoyu Jiang,Yilun Lin.CredID: Credible Multi-Bit Watermark for Large Language Models Identification[EB/OL].(2024-12-04)[2025-08-02].https://arxiv.org/abs/2412.03107.点此复制

评论