GGBall: Graph Generative Model on Poincar\'e Ball

TL;DR

GGBall introduces a hyperbolic graph generative model that effectively captures hierarchical structures using geometric inductive biases, Riemannian flow matching, and manifold-based neural layers, significantly improving topological hierarchy preservation.

Contribution

It presents GGBall, a novel hyperbolic framework combining HVQVAE, Riemannian flow, and manifold neural layers for graph generation, addressing limitations of Euclidean geometry.

Findings

Reduces degree MMD by over 75% on Community-Small

Reduces degree MMD by over 40% on Ego-Small

Improves preservation of topological hierarchies

Abstract

Generating graphs with hierarchical structures remains a fundamental challenge due to the limitations of Euclidean geometry in capturing exponential complexity. Here we introduce \textbf{GGBall}, a novel hyperbolic framework for graph generation that integrates geometric inductive biases with modern generative paradigms. GGBall combines a Hyperbolic Vector-Quantized Autoencoder (HVQVAE) with a Riemannian flow matching prior defined via closed-form geodesics. This design enables flow-based priors to model complex latent distributions, while vector quantization helps preserve the curvature-aware structure of the hyperbolic space. We further develop a suite of hyperbolic GNN and Transformer layers that operate entirely within the manifold, ensuring stability and scalability. Empirically, our model reduces degree MMD by over 75\% on Community-Small and over 40\% on Ego-Small compared to state-of-the-art baselines, demonstrating an improved ability to preserve topological hierarchies. These results highlight the potential of hyperbolic geometry as a powerful foundation for the generative modeling of complex, structured, and hierarchical data domains. Our code is available at \href{https://github.com/AI4Science-WestlakeU/GGBall}{here}.