Search papers, labs, and topics across Lattice.
This paper introduces PriDyG, a privacy-preserving framework for dynamic graph inference that integrates Graph Neural Networks (GNNs) with Large Language Models (LLMs) to mitigate privacy risks associated with sensitive edge information. By employing Edge-level Differentially Private Dynamic Graph Inference (EDG) and incremental private multi-hop aggregation, PriDyG processes each edge only once while maintaining a low privacy cost, independent of the number of model updates. Experimental results demonstrate that PriDyG significantly outperforms traditional methods, reducing cumulative privacy costs by up to three orders of magnitude while preserving utility in node classification and link prediction tasks.
PriDyG slashes cumulative privacy costs by up to 99.9% while maintaining high utility in dynamic graph inference.
Graph inference over relational data can expose sensitive edge information, and this risk becomes more severe in dynamic graphs, where repeated model updates cause privacy loss to accumulate. We formulate Edge-level Differentially Private Dynamic Graph Inference (EDG) and propose PriDyG, a private inference framework that combines GNN-based structural learning with LLM-based semantic reasoning. PriDyG introduces incremental private multi-hop aggregation, which buffers newly arrived edges and processes each edge exactly once. By parallel composition, the total privacy cost equals that of a single static release, independent of the number or schedule of model updates. Compared with geometrically decaying budget allocation, incremental aggregation avoids exponentially increasing noise while preserving exact one-hop signals and at least half of two-hop information transfers. PriDyG further complements privatized GNN outputs with LLM predictions derived solely from node text, incurring no additional edge-level privacy cost. Experiments on four benchmarks for node classification and link prediction show that PriDyG consistently outperforms geometrically decaying baselines under the same privacy budget and matches the utility of naive per-update retraining while reducing cumulative privacy cost by up to three orders of magnitude.
