From Trade-off to Synergy: A Versatile Symbiotic Watermarking Framework for Large Language Models

TL;DR

This paper introduces a versatile watermarking framework for LLMs that combines logits-based and sampling-based methods to improve robustness, security, and text quality, validated by extensive experiments.

Contribution

It proposes a hybrid watermarking framework with adaptive strategies that outperform existing methods and achieve state-of-the-art results.

Findings

Outperforms existing watermarking baselines.

Achieves state-of-the-art robustness and security.

Maintains high text quality.

Abstract

The rise of Large Language Models (LLMs) has heightened concerns about the misuse of AI-generated text, making watermarking a promising solution. Mainstream watermarking schemes for LLMs fall into two categories: logits-based and sampling-based. However, current schemes entail trade-offs among robustness, text quality, and security. To mitigate this, we integrate logits-based and sampling-based schemes, harnessing their respective strengths to achieve synergy. In this paper, we propose a versatile symbiotic watermarking framework with three strategies: serial, parallel, and hybrid. The hybrid framework adaptively embeds watermarks using token entropy and semantic entropy, optimizing the balance between detectability, robustness, text quality, and security. Furthermore, we validate our approach through comprehensive experiments on various datasets and models. Experimental results indicate that our method outperforms existing baselines and achieves state-of-the-art (SOTA) performance. We believe this framework provides novel insights into diverse watermarking paradigms. Our code is available at https://github.com/redwyd/SymMark.