EUVIMEDEUVIMED entdecken
Crossref

QRShield: Exploiting Vulnerabilities of Latent Diffusion Models for Preventing AI Art Plagiarism

Xunyue Mo, Weibin Wu, Qingrui Tu, Hang Wang, Junxi He, Zibin Zheng

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (10) · 8098-8106

DOI: 10.1609/aaai.v40i10.37756 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Xunyue Mo und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Latent Diffusion Models (LDMs) have achieved remarkable success in image generation tasks, yet their low barrier to customization poses severe threats related to art plagiarism. As a countermeasure, adversarial methods have been proposed to protect artworks from plagiarism. However, current methods suffer from limited effectiveness, high cost, and complex optimization. Moreover, their exploration and exploitation of LDM vulnerabilities remain limited, restricting effectiveness and applicability. To address this issue, we analyze the VAE and U-Net components of LDMs, revealing their vulnerabilities. Specifically, we study the response of U-Net to specific structural and frequency patterns in the latent space and find that it is susceptible to high-frequency and periodic latent features. Furthermore, we observe channel correlations during the VAE encoding process. Inspired by these, we propose QRShield, an efficient protection method that exploits the vulnerabilities of LDMs. By constructing high-frequency and periodic features consistent across latent channels and combining them with a momentum-based translation-invariant attack strategy, QRShield achieves stronger and more efficient protection. QRShield significantly improves protection performance in various fine-tuning settings, with over 10% gains in multiple metrics, a threefold increase in generation speed, and nearly 50% reduction in memory usage. Therefore, our work offers a more practical method to prevent AI art plagiarism.

Zitieren

APA

Mo, X., Wu, W., Tu, Q., Wang, H., He, J., & Zheng, Z. (2026). QRShield: Exploiting Vulnerabilities of Latent Diffusion Models for Preventing AI Art Plagiarism. Proceedings of the AAAI Conference on Artificial Intelligence, 40(10), 8098-8106. https://doi.org/10.1609/aaai.v40i10.37756

Vancouver

Mo X, Wu W, Tu Q, Wang H, He J, Zheng Z. QRShield: Exploiting Vulnerabilities of Latent Diffusion Models for Preventing AI Art Plagiarism. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(10):8098-8106. doi:10.1609/aaai.v40i10.37756

BibTeX
@article{Mo2026,
  author = {Mo, Xunyue and Wu, Weibin and Tu, Qingrui and Wang, Hang and He, Junxi and Zheng, Zibin},
  title = {QRShield: Exploiting Vulnerabilities of Latent Diffusion Models for Preventing AI Art Plagiarism},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {10},
  pages = {8098--8106},
  doi = {10.1609/aaai.v40i10.37756}
}
Kostenlos starten →