Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework

Yin, Zhipeng; Wang, Zichong; Chen, Ruijun; Ning, Xin; Zhang, Xingyu; Zhang, Wenbin

doi:10.24963/ijcai.2026/96

Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework

Zhipeng Yin, Zichong Wang, Ruijun Chen, Xin Ning, Xingyu Zhang, Wenbin Zhang

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 853-861.

https://doi.org/10.24963/ijcai.2026/96

Keywords

AI Ethics, Trust, Fairnes: AI and law, governance, regulation
AI Ethics, Trust, Fairnes: Societal impact of AI
AI Ethics, Trust, Fairnes: Ethical, legal and societal issues
AI Ethics, Trust, Fairnes: Trustworthy AI

Abstract

Text-to-Image (T2I) diffusion models have been widely adopted due to their strong generative capabilities, while Low-Rank Adaptation (LoRA) has emerged as an efficient mechanism for customizing these models for diverse creative and commercial applications. This trend has fostered LoRA-centric service platforms that that enable the customization and commercial distribution of LoRA modules according to user requirements. However, the growing prevalence of LoRA and its critical role in customized AI services have raised urgent concerns about LoRA copyright protection. To address this gap, we propose LoRA2D, an authorized dual-watermarking framework specifically designed to protect LoRA modules in T2I diffusion models. LoRA2D integrates license-based authorization control with explicit watermarks as visible deterrents for unauthorized or trial usage, which can be removed upon valid authorization, while persistently embedding an implicit watermark for robust black-box ownership verification. Extensive experiments on multiple image-generation datasets demonstrate the effectiveness and practicality of LoRA2D for securing copyrights in LoRA-adapted T2I diffusion models.

Zhipeng Yin, Zichong Wang, Ruijun Chen, Xin Ning, Xingyu Zhang, and Wenbin Zhang. 2026. Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 853–861. https://doi.org/10.24963/ijcai.2026/96

@inproceedings{ijcai2026p96,
  title     = {Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework},
  author    = {Yin, Zhipeng and Wang, Zichong and Chen, Ruijun and Ning, Xin and Zhang, Xingyu and Zhang, Wenbin},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {853--861},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/96},
  url       = {https://doi.org/10.24963/ijcai.2026/96},
}

TY  - CPAPER
TI  - Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework
AU  - Yin, Zhipeng
AU  - Wang, Zichong
AU  - Chen, Ruijun
AU  - Ning, Xin
AU  - Zhang, Xingyu
AU  - Zhang, Wenbin
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 853
EP  - 861
DO  - 10.24963/ijcai.2026/96
UR  - https://doi.org/10.24963/ijcai.2026/96
ER  -