In-Context Watermarks for Large Language Models

TL;DR

This paper proposes In-Context Watermarking (ICW), a prompt-based method for embedding watermarks into LLM-generated text without needing access to the decoding process, enhancing practicality and scalability.

Contribution

The paper introduces ICW, a novel prompt engineering approach for watermarking LLM outputs, applicable in real-world scenarios without model access, and demonstrates its effectiveness through experiments.

Findings

ICW is feasible as a model-agnostic watermarking technique.

Different ICW strategies can be paired with tailored detection methods.

ICW shows promise for scalable content attribution as LLMs advance.

Abstract

The growing use of large language models (LLMs) for sensitive applications has highlighted the need for effective watermarking techniques to ensure the provenance and accountability of AI-generated text. However, most existing watermarking methods require access to the decoding process, limiting their applicability in real-world settings. One illustrative example is the use of LLMs by dishonest reviewers in the context of academic peer review, where conference organizers have no access to the model used but still need to detect AI-generated reviews. Motivated by this gap, we introduce In-Context Watermarking (ICW), which embeds watermarks into generated text solely through prompt engineering, leveraging LLMs' in-context learning and instruction-following abilities. We investigate four ICW strategies at different levels of granularity, each paired with a tailored detection method. We further examine the Indirect Prompt Injection (IPI) setting as a specific case study, in which watermarking is covertly triggered by modifying input documents such as academic manuscripts. Our experiments validate the feasibility of ICW as a model-agnostic, practical watermarking approach. Moreover, our findings suggest that as LLMs become more capable, ICW offers a promising direction for scalable and accessible content attribution. Our code is available at https://github.com/yepengliu/In-Context-Watermarks.