Journal article · 2026 · Open access
Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation
IEEE Access, 14, 9319–9339 · Published 1 January 2026 · DOI 10.1109/access.2026.3653833
In brief
“Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation” is a 2026 journal article by Ziyang Zeng and 3 co-authors, published in IEEE Access. OpenAlex counts 33 citations as of 18 November 2026, which puts it in the top 1% of works published in 2026 by citations. It is open access under a CC BY licence: the full text is free to read at IEEE Access.
- Citations
- 33
- Most cited in
- 2026 (33)
- Field-weighted impact
- 37.75
- References
- 40
- Authors
- 4
Abstract
Large Language Models (LLMs) have demonstrated remarkable capabilities in generating high-quality text, raising significant concerns regarding copyright protection and content provenance verification. However, most existing watermarking techniques rely on uniform perturbation or rule-based token biasing schemes, which exhibit critical vulnerabilities under adversarial attacks such as paraphrasing, translation, and content truncation, often failing to maintain detection reliability in real-world deployment scenarios. To address these challenges, this paper introduces a novelcontext-aware robust watermarking frameworkthat dynamically adjusts watermark embedding strength according to contextual semantic characteristics during text generation. The proposed approach incorporates a token-level semantic modulation mechanism that strategically intensifies watermark signals in copyright-sensitive segments while minimizing perturbations in semantically neutral regions, achieving an improved balance between imperceptibility and robustness. Furthermore, an adaptive threshold estimation algorithm is developed for watermark detection, which automatically calibrates detection boundaries based on noise statistics, significantly enhancing resilience against diverse attack vectors. Extensive experiments on the WaterBench benchmark demonstrate superior performance over state-of-the-art baselines, maintaining high detection accuracy with a 95.3% true positive rate (TPR) under clean conditions and strong robustness under severe perturbations, including paraphrasing attacks (82.7% TPR), translation attacks (78.4% TPR), and content truncation (88.9% TPR at 50% retention). Meanwhile, the proposed method reduces false positive rates by 43.2% compared with existing approaches while preserving text quality with negligible perplexity increase (1.8%). These results establish a new paradigm for practical and scalable LLM watermarking in real-world copyright-sensitive deployment scenarios.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Zeng, Z., Lin, H., Zhang, S., & Wang, B. (2026). Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation. IEEE Access, 14, 9319–9339. https://doi.org/10.1109/access.2026.3653833
MLA 9
Zeng, Ziyang, et al. “Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation.” IEEE Access, vol. 14, 2026, pp. 9319–9339. https://doi.org/10.1109/access.2026.3653833.
Chicago
Zeng, Ziyang, Han Lin, Shuxin Zhang, and Boyuan Wang. 2026. “Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation.” IEEE Access 14: 9319–9339. https://doi.org/10.1109/access.2026.3653833.
Harvard
Zeng, Z. et al. (2026) ‘Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation’, IEEE Access, 14, pp. 9319–9339. doi:10.1109/access.2026.3653833.
IEEE
Z. Zeng, H. Lin, S. Zhang, and B. Wang, “Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation,” IEEE Access, vol. 14, pp. 9319–9339, 2026, doi: 10.1109/access.2026.3653833.
BibTeX
@article{zeng2026adaptive,
title = {{Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation}},
author = {Zeng, Ziyang and Lin, Han and Zhang, Shuxin and Wang, Boyuan},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {9319--9339},
doi = {10.1109/access.2026.3653833},
url = {https://doi.org/10.1109/access.2026.3653833}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Steganography and Watermarking Techniques
- copyright protection
- semantic modulation
- paraphrasing attacks
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Zeng, Z., Lin, H., Zhang, S., & Wang, B. (2026). Adaptive Robust Watermarking for Large Language Models via Dynamic Token Embedding Perturbation. IEEE Access, 14, 9319–9339. https://doi.org/10.1109/access.2026.3653833
How many times has this paper been cited?
33 times, according to OpenAlex on 18 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at IEEE Access: sign in and open the full text from this page.
Who wrote this paper?
Ziyang Zeng, Han Lin, Shuxin Zhang, Boyuan Wang.
What is the DOI of this paper?
10.1109/access.2026.3653833, permanently at https://doi.org/10.1109/access.2026.3653833. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 18 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.