MorphMark: Flexible Adaptive Watermarking for Large Language Models

TL;DR

MorphMark introduces an adaptive, model-agnostic watermarking method for large language models that balances effectiveness and text quality, overcoming a key trade-off in existing techniques.

Contribution

It develops MorphMark, a novel watermarking approach that adaptively adjusts watermark strength based on a formalized trade-off framework, enhancing practicality and flexibility.

Findings

Achieves better effectiveness-quality balance

Demonstrates superior flexibility and efficiency

Outperforms existing watermarking methods

Abstract

Watermarking by altering token sampling probabilities based on red-green list is a promising method for tracing the origin of text generated by large language models (LLMs). However, existing watermark methods often struggle with a fundamental dilemma: improving watermark effectiveness (the detectability of the watermark) often comes at the cost of reduced text quality. This trade-off limits their practical application. To address this challenge, we first formalize the problem within a multi-objective trade-off analysis framework. Within this framework, we identify a key factor that influences the dilemma. Unlike existing methods, where watermark strength is typically treated as a fixed hyperparameter, our theoretical insights lead to the development of MorphMarka method that adaptively adjusts the watermark strength in response to changes in the identified factor, thereby achieving an effective resolution of the dilemma. In addition, MorphMark also prioritizes flexibility since it is a model-agnostic and model-free watermark method, thereby offering a practical solution for real-world deployment, particularly in light of the rapid evolution of AI models. Extensive experiments demonstrate that MorphMark achieves a superior resolution of the effectiveness-quality dilemma, while also offering greater flexibility and time and space efficiency.