KGMark: A Diffusion Watermark for Knowledge Graphs

Hongrui Peng, Haolang Lu, Yuanlong Yu, WeiYe Fu, Kun Wang, Guoshun Nan

TL;DR: We present KGMark, the first watermarking method for knowledge graph embeddings that ensures high detectability, transparency, and robustness across various graph modifications.

Abstract: 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.

Lay Summary: Knowledge graphs are widely used to represent structured information, but verifying their ownership is hard once shared online. We developed a watermarking method that hides invisible signatures inside the graph's learned embeddings using a diffusion model. Our method resists both random and targeted tampering, helping protect graph-based AI assets from unauthorized use and plagiarism.

Primary Area: Social Aspects->Fairness

Keywords: Watermarking, Knowledge Graph, Diffusion Models, Generative Models

Submission Number: 16409