Skip to main navigation Skip to search Skip to main content

Learning to Watermark: A Selective Watermarking Framework for Large Language Models via Multi-Objective Optimization

  • Chenrui WANG
  • , Junyi SHU
  • , Billy CHIU
  • , Yu LI
  • , Saleh ALHARBI
  • , Min ZHANG
  • , Jing LI*
  • *Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

Abstract

The rapid development of LLMs has raised concerns about their potential misuse, leading to various watermarking schemes that typically offer high detectability. However, existing watermarking techniques often face trade-off between watermark detectability and generated text quality. In this paper, we introduce Learning to Watermark (LTW), a novel selective watermarking framework that leverages multi-objective optimization to effectively balance these competing goals. LTW features a lightweight network that adaptively decides when to apply the watermark by analyzing sentence embeddings, token entropy, and current watermarking ratio. Training of the network involves two specifically constructed loss functions that guide the model toward Pareto-optimal solutions, thereby harmonizing watermark detectability and text quality. By integrating LTW with two baseline watermarking methods, our experimental evaluations demonstrate that LTW significantly enhances text quality without compromising detectability. Our selective watermarking approach offers a new perspective for designing watermarks for LLMs and a way to preserve high text quality for watermarks. The code is publicly available at: https://github.com/fattyray/learning-to-watermark.

Original languageEnglish
Title of host publication39th Conference on Neural Information Processing Systems, NeurIPS 2025: Proceedings
PublisherNeural Information Processing Systems Foundation
Pages143745-143772
Number of pages28
ISBN (Electronic)9798331338275
DOIs
Publication statusPublished - Dec 2025
Event39th Conference on Neural Information Processing Systems, NeurIPS 2025 - San Diego, United States
Duration: 2 Dec 2025 → 7 Dec 2025

Publication series

NameAdvances in Neural Information Processing Systems
Volume38
ISSN (Print)1049-5258

Conference

Conference39th Conference on Neural Information Processing Systems, NeurIPS 2025
Country/TerritoryUnited States
CitySan Diego
Period2/12/25 → 7/12/25

Bibliographical note

Publisher Copyright:
© 2025 Neural information processing systems foundation. All rights reserved.

Funding

This work was supported in part by National Natural Science Foundation of China (62476070), Shenzhen Science and Technology Program (JCYJ20241202123503005, GXWD20231128103232001, ZDSYS20230626091203008, KQTD20240729102154066), Department of Science and Technology of Guangdong (2024A1515011540) and National Key R&D Program of China (SQ2024YFE0200592).

Fingerprint

Dive into the research topics of 'Learning to Watermark: A Selective Watermarking Framework for Large Language Models via Multi-Objective Optimization'. Together they form a unique fingerprint.

Cite this