KGMark: A Diffusion Watermark for Knowledge Graphs
This work addresses the need to protect intellectual property and prevent harm from AI-generated content in knowledge graphs, which are widely used in real-world applications, representing a novel domain-specific advancement.
The authors tackled the problem of watermarking dynamic knowledge graphs, which existing methods could not handle due to spatial and temporal variations, and proposed KGMark, a framework that generates robust, detectable, and transparent diffusion fingerprints, showing effectiveness in experiments on public benchmarks.
Knowledge graphs (KGs) are ubiquitous in numerous real-world applications, and watermarking facilitates protecting intellectual property and preventing potential harm from AI-generated content. Existing watermarking methods mainly focus on static plain text or image data, while they can hardly be applied to dynamic graphs due to spatial and temporal variations of structured data. This motivates us to propose KGMARK, the first graph watermarking framework that aims to generate robust, detectable, and transparent diffusion fingerprints for dynamic KG data. Specifically, we propose a novel clustering-based alignment method to adapt the watermark to spatial variations. Meanwhile, we present a redundant embedding strategy to harden the diffusion watermark against various attacks, facilitating the robustness of the watermark to the temporal variations. Additionally, we introduce a novel learnable mask matrix to improve the transparency of diffusion fingerprints. By doing so, our KGMARK properly tackles the variation challenges of structured data. Experiments on various public benchmarks show the effectiveness of our proposed KGMARK. Our code is available at https://github.com/phrara/kgmark.