An Entropy-based Text Watermarking Detection Method

TL;DR

This paper introduces EWD, an entropy-aware watermark detection method for LLM-generated texts that improves detection in low-entropy scenarios by weighting tokens according to their entropy, without requiring training.

Contribution

The paper proposes a novel entropy-based weighting scheme for watermark detection that enhances performance in low-entropy contexts and is fully automated and training-free.

Findings

EWD outperforms previous methods in low-entropy scenarios.

The method is general and adaptable to texts with varying entropy.

EWD is training-free and fully automated.

Abstract

Text watermarking algorithms for large language models (LLMs) can effectively identify machine-generated texts by embedding and detecting hidden features in the text. Although the current text watermarking algorithms perform well in most high-entropy scenarios, its performance in low-entropy scenarios still needs to be improved. In this work, we opine that the influence of token entropy should be fully considered in the watermark detection process, $i.e.$, the weight of each token during watermark detection should be customized according to its entropy, rather than setting the weights of all tokens to the same value as in previous methods. Specifically, we propose \textbf{E}ntropy-based Text \textbf{W}atermarking \textbf{D}etection (\textbf{EWD}) that gives higher-entropy tokens higher influence weights during watermark detection, so as to better reflect the degree of watermarking. Furthermore, the proposed detection process is training-free and fully automated. From the experiments, we demonstrate that our EWD can achieve better detection performance in low-entropy scenarios, and our method is also general and can be applied to texts with different entropy distributions. Our code and data is available\footnote{\url{https://github.com/luyijian3/EWD}}. Additionally, our algorithm could be accessed through MarkLLM \cite{pan2024markllm}\footnote{\url{https://github.com/THU-BPM/MarkLLM}}.