Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach

Haiyun He, Yepeng Liu, Ziqiao Wang, Yongyi Mao, Yuheng Bu

Track: long paper (up to 9 pages)

Keywords: LLM, Watermark, Theory, Distribution-Adaptive, Hypothesis Testing, Trustworthy

Abstract: Watermarking has emerged as a crucial method to distinguish AI-generated text from human-created text. In this paper, we present a novel theoretical framework for watermarking Large Language Models (LLMs) that jointly optimizes both the watermarking scheme and the detection process. Our approach focuses on maximizing detection performance while maintaining control over the worst-case Type-I error and text distortion. We characterize \emph{the universally minimum Type-II error}, showing a fundamental trade-off between watermark detectability and text distortion. Importantly, we identify that the optimal watermarking schemes are adaptive to the LLM generative distribution. Building on our theoretical insights, we propose an efficient, model-agnostic, distribution-adaptive watermarking algorithm, utilizing a surrogate model alongside the Gumbel-max trick. Experiments conducted on Llama2-13B and Mistral-8$\times$7B models confirm the effectiveness of our approach. Additionally, we examine incorporating robustness into our framework, paving the way for future watermarking systems that withstand adversarial attacks more effectively.

Presenter: ~Yepeng_Liu1

Format: Yes, the presenting author will attend in person if this work is accepted to the workshop.

Funding: Yes, the presenting author of this submission falls under ICLR’s funding aims, and funding would significantly impact their ability to attend the workshop in person.

Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.

Submission Number: 31