PARSE：部件感知的关系空间建模
PARSE: Part-Aware Relational Spatial Modeling
提出部件级装配图（PAG）作为结构化先验，可注入3D生成模型提升物理真实感与结构复杂度；同时为视觉语言模型提供部件级关系监督，增强空间推理能力。
原文依据 leveraging PAGs as structural priors in 3D generation models leads to scenes with substantially improved physical realism and structural complexity
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论物体间关系是空间智能的基础，然而现有表示——语言介词或物体级场景图——过于粗糙，无法指明哪些区域实际支撑、包含或接触彼此，导致布局模糊且物理不一致。为解决这些歧义，需要部件级公式化表示；因此，我们提出PARSE，一个显式建模物体部件如何交互以确定可行且空间接地的场景配置的框架。PARSE以部件中心装配图（PAG）为核心，该图编码特定物体部件间的几何关系，并采用部件感知空间配置求解器将这些关系转化为几何约束，以组装无碰撞、物理有效的场景。利用PARSE，我们构建了PARSE-10K数据集，包含10,000个3D室内场景，由真实图像布局先验和精选的部件标注形状数据库构建，每个场景具有密集接触结构和部件级接触图。借助这种结构化、空间接地的监督，在PARSE-10K上微调Qwen3-VL可获得更强的物体级布局推理和更准确的部件级关系理解；此外，将PAG作为结构先验用于3D生成模型，可生成物理真实感和结构复杂度显著提升的场景。这些结果共同表明，PARSE显著推进了几何接地的空间推理，并支持生成物理一致的3D场景。
英文原文摘要 1524 字符
Inter-object relations underpin spatial intelligence, yet existing representations--linguistic prepositions or object-level scene graphs--are too coarse to specify which regions actually support, contain, or contact one another, leading to ambiguous and physically inconsistent layouts. To address these ambiguities, a part-level formulation is needed; therefore, we introduce PARSE, a framework that explicitly models how object parts interact to determine feasible and spatially grounded scene configurations. PARSE centers on the Part-centric Assembly Graph (PAG), which encodes geometric relations between specific object parts, and a Part-Aware Spatial Configuration Solver that converts these relations into geometric constraints to assemble collision-free, physically valid scenes. Using PARSE, we build PARSE-10K, a dataset of 10,000 3D indoor scenes constructed from real-image layout priors and a curated part-annotated shape database, each with dense contact structures and a part-level contact graph. With this structured, spatially grounded supervision, fine-tuning Qwen3-VL on PARSE-10K yields stronger object-level layout reasoning and more accurate part-level relation understanding; furthermore, leveraging PAGs as structural priors in 3D generation models leads to scenes with substantially improved physical realism and structural complexity. Together, these results show that PARSE significantly advances geometry-grounded spatial reasoning and supports the generation of physically consistent 3D scenes.