Watermarking Generative models

Generative models are increasingly used to create synthetic data across domains, from text and images to tabular data and time series. However, the proliferation of AI-generated content raises critical concerns about model ownership, data provenance, and intellectual property protection. Watermarking techniques offer a solution by embedding verifiable signatures into generative content, enabling creators to prove authenticity and trace misuse while maintaining the utility and quality of generated content.

Publications

CheckMate! Watermarking Graph Diffusion Models in Polynomial Time

Roberto Gheda, Abele Mălan, Robert Birke, Maksim Kitsak, Lydia Y. Chen

ICLR 2026 : 📄 Paper | 💻 Code

Citation

@inproceedings{GhedaICLR26CheckMate,
  author    = {Roberto Gheda and
     Abele Mălan and
     Robert Birke and
     Maksim Kitsak and
     Lydia Y. Chen },
  title     = {CheckMate! Watermarking Graph Diffusion Models in Polynomial Time},
  booktitle = {ICLR },
  year      = {2026}
}

TimeWak: Temporal Chained-Hashing Watermark for Time Series Data

Zhi Wen Soi, Chaoyi Zhu, Fouad Abiad, Aditya Shankar, Jeroen M. Galjaard, Huijuan Wang, Lydia Y. Chen (*equal contribution)

NeurIPS 2025 (🏆Spotlight): 📄Paper|💻 Code

Citation

@article{DBLP:journals/corr/abs-2506-06407,
  author       = {Zhi Wen Soi and
                  Chaoyi Zhu and
                  Fouad Abiad and
                  Aditya Shankar and
                  Jeroen M. Galjaard and
                  Huijuan Wang and
                  Lydia Y. Chen},
  title        = {TimeWak: Temporal Chained-Hashing Watermark for Time Series Data},
  journal      = {CoRR},
  volume       = {abs/2506.06407},
  year         = {2025},
  url          = {https://doi.org/10.48550/arXiv.2506.06407},
  doi          = {10.48550/ARXIV.2506.06407},
  eprinttype    = {arXiv},
  eprint       = {2506.06407},
  timestamp    = {Mon, 07 Jul 2025 21:56:27 +0200},
  biburl       = {https://dblp.org/rec/journals/corr/abs-2506-06407.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

TabWak: A Watermark for Tabular Diffusion Models

Chaoyi Zhu, Jiayi Tang, Jeroen M. Galjaard, Pin-Yu Chen, Robert Birke, Cornelis Bos, Lydia Y. Chen

ICLR 2025 (🏆 Spotlight): 📄Paper|💻 Code| ��️Poster

Citation

@inproceedings{DBLP:conf/iclr/ZhuTGCBBC25,
  author       = {Chaoyi Zhu and
                  Jiayi Tang and
                  Jeroen M. Galjaard and
                  Pin{-}Yu Chen and
                  Robert Birke and
                  Cornelis Bos and
                  Lydia Y. Chen},
  title        = {TabWak: {A} Watermark for Tabular Diffusion Models},
  booktitle    = {The Thirteenth International Conference on Learning Representations,
                  {ICLR} 2025, Singapore, April 24-28, 2025},
  publisher    = {OpenReview.net},
  year         = {2025},
  url          = {https://openreview.net/forum?id=71pur4y8gs},
  timestamp    = {Thu, 15 May 2025 17:19:05 +0200},
  biburl       = {https://dblp.org/rec/conf/iclr/ZhuTGCBBC25.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

Duwak: Dual Watermarks in Large Language Models

Chaoyi Zhu, Jeroen Galjaard, Pin-Yu Chen, Lydia Y. Chen

ACL 2024 📄 Paper|💻 Code

Citation

@inproceedings{DBLP:conf/acl/ZhuGCC24,
  author       = {Chaoyi Zhu and
                  Jeroen Galjaard and
                  Pin{-}Yu Chen and
                  Lydia Y. Chen},
  editor       = {Lun{-}Wei Ku and
                  Andre Martins and
                  Vivek Srikumar},
  title        = {Duwak: Dual Watermarks in Large Language Models},
  booktitle    = {Findings of the Association for Computational Linguistics, {ACL} 2024,
                  Bangkok, Thailand and virtual meeting, August 11-16, 2024},
  pages        = {11416--11436},
  publisher    = {Association for Computational Linguistics},
  year         = {2024},
  url          = {https://doi.org/10.18653/v1/2024.findings-acl.678},
  doi          = {10.18653/V1/2024.FINDINGS-ACL.678},
  timestamp    = {Tue, 24 Sep 2024 10:55:44 +0200},
  biburl       = {https://dblp.org/rec/conf/acl/ZhuGCC24.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

Optimization-Free Universal Watermark Forgery with Regenerative Diffusion Models

Chaoyi Zhu, Zaitang Li, Renyi Yang, Robert Birke, Pin-Yu Chen, Tsung-Yi Ho, Lydia Y. Chen

📄 Paper

Citation

@misc{zhu2025optimizationfreeuniversalwatermarkforgery,
      title={Optimization-Free Universal Watermark Forgery with Regenerative Diffusion Models}, 
      author={Chaoyi Zhu and Zaitang Li and Renyi Yang and Robert Birke and Pin-Yu Chen and Tsung-Yi Ho and Lydia Y. Chen},
      year={2025},
      eprint={2506.06018},
      archivePrefix={arXiv},
      primaryClass={cs.MM},
      url={https://arxiv.org/abs/2506.06018}, 
}