Graph generative models are essential across diverse scientific domains by capturing complex distributions over relational data. Among them, graph diffusion models achieve superior performance but face inefficient sampling and limited flexibility due to the tight coupling between training and sampling stages. We introduce DeFoG, a novel graph generative framework that disentangles sampling from training, enabling a broader design space for more effective and efficient model optimization. DeFoG employs a discrete flow-matching formulation that respects the inherent symmetries of graphs. We theoretically ground this disentangled formulation by explicitly relating the training loss to the sampling algorithm and showing that DeFoG faithfully replicates the ground truth graph distribution. Building on these foundations, we thoroughly investigate DeFoG's design space and propose novel sampling methods that significantly enhance performance and reduce the required number of refinement steps. Extensive experiments demonstrate state-of-the-art performance across synthetic, molecular, and digital pathology datasets, covering both unconditional and conditional generation settings. It also outperforms most diffusion-based models with just 5-10% of their sampling steps.

翻译：图生成模型通过捕捉关系数据上的复杂分布，在多个科学领域中具有关键作用。其中，图扩散模型虽取得了卓越性能，但由于训练与采样阶段的紧密耦合，面临采样效率低下和灵活性受限的问题。我们提出了DeFoG，一种新颖的图生成框架，它将采样过程与训练解耦，从而为更高效、更有效的模型优化提供了更广阔的设计空间。DeFoG采用了一种尊重图固有对称性的离散流匹配形式。我们从理论上为该解耦形式建立了基础，通过明确地将训练损失与采样算法联系起来，并证明DeFoG能够忠实复现真实图分布。基于此，我们深入探索了DeFoG的设计空间，并提出了新颖的采样方法，这些方法显著提升了性能并减少了所需的细化步骤数量。大量实验表明，在合成、分子和数字病理学数据集上，无论是无条件生成还是有条件生成任务，DeFoG均实现了最先进的性能。此外，它仅需基于扩散模型5-10%的采样步骤，即可超越大多数此类模型。