Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework

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

IJCAI 20260 citations

Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework

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

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.

AI Ethics, Trust, Fairnes: AI and law, governance, regulationAI Ethics, Trust, Fairnes: Ethical, legal and societal issuesAI Ethics, Trust, Fairnes: Societal impact of AIAI Ethics, Trust, Fairnes: Trustworthy AI

BibTeX

@inproceedings{ijcai2026_towardloracopyri,
  title = {Toward LoRA Copyright Protection with an Authorized Dual-Watermarking Framework},
  author = {Zhipeng Yin and Zichong Wang and Ruijun Chen and Xin Ning and Xingyu Zhang and Wenbin Zhang},
  booktitle = {IJCAI 2026},
  year = {2026}
}