Abstract
Compositionality is critical for 3D object and scene generation, but existing part-aware 3D generation methods suffer from poor scalability due to quadratic global attention costs when increasing the number of components. In this work, we present MoCA, a compositional 3D generative model with two key designs: (1) importance-based component routing that selects top-k relevant components for sparse global attention, and (2) unimportant components compression that preserve contextual priors of unselected components while reducing computational complexity of global attention. With these designs, MoCA enables efficient, fine-grained compositional 3D asset creation with scalable number of components. Extensive experiments show MoCA outperforms baselines on both compositional object and scene generation tasks. Project page: this https URL
Abstract (translated)
组合性对于生成三维物体和场景至关重要，但现有的具有部件感知能力的3D生成方法在增加组件数量时由于二次全局注意力成本而面临扩展性的挑战。在这项工作中，我们提出了MoCA，这是一种组合式3D生成模型，具备两个关键设计：(1) 基于重要性进行组件路由，选择前k个相关的组件用于稀疏的全局注意力；(2) 对不重要的组件进行压缩，在减少全局注意力计算复杂度的同时保留未选组件的上下文先验。通过这两个设计，MoCA能够实现高效、细粒度的组合式3D资产创建，并且支持可扩展数量的组件。大量的实验表明，MoCA在组合式物体和场景生成任务上超越了基准方法。项目页面：[此链接](this https URL)
URL
https://arxiv.org/abs/2512.07628
