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

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 conduct an in-depth analysis of the VAE and U-Net components within LDMs, revealing their inherent 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 significant 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 consistent high-frequency and periodic features 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. Our work deeply reveals the vulnerabilities of LDMs and proposes a more practical tool to prevent AI art plagiarism.