SpatialStack：面向3D VLM空间推理的分层几何-语言融合
SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning
把多视图几何编码器的多层特征逐级同步注入语言主干，替代只在深层做视觉-几何晚期融合的做法；做3D VLM或需要精细空间关系理解的工作可直接复用这一分层融合范式。
原文依据 SpatialStack stacks and synchronizes multi-level geometric features with the language backbone
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论大型视觉语言模型（VLM）在可靠的3D空间推理上仍存在困难，而这是具身智能与物理AI系统的核心能力。这一局限源于其无法捕捉细粒度的3D几何与空间关系。尽管近期工作已将多视图几何Transformer引入VLM，但它们通常只融合视觉编码器与几何编码器的深层特征，丢弃了丰富的层次化信号，从而为空间理解制造了根本性瓶颈。为克服这一问题，我们提出SpatialStack，一种通用的分层融合框架，在模型各层级上逐步对齐视觉、几何与语言表征。SpatialStack超越了传统的视觉-几何晚期融合，将多层级几何特征与语言主干堆叠并同步，使模型能够同时捕捉局部几何精度与全局上下文语义。在该框架基础上，我们开发了VLM-SpatialStack模型，在多个3D空间推理基准上取得最优性能。大量实验与消融表明，我们的多层级融合策略持续增强3D理解，并在多样的空间推理任务上稳健泛化，确立了SpatialStack作为下一代多模态物理AI系统中视觉-语言-几何集成的有效且可扩展的设计范式。
英文原文摘要 1461 字符
Large vision-language models (VLMs) still struggle with reliable 3D spatial reasoning, a core capability for embodied and physical AI systems. This limitation arises from their inability to capture fine-grained 3D geometry and spatial relationships. While recent efforts have introduced multi-view geometry transformers into VLMs, they typically fuse only the deep-layer features from vision and geometry encoders, discarding rich hierarchical signals and creating a fundamental bottleneck for spatial understanding. To overcome this, we propose SpatialStack, a general hierarchical fusion framework that progressively aligns vision, geometry, and language representations across the model hierarchy. Moving beyond conventional late-stage vision-geometry fusion, SpatialStack stacks and synchronizes multi-level geometric features with the language backbone, enabling the model to capture both local geometric precision and global contextual semantics. Building upon this framework, we develop VLM-SpatialStack, a model that achieves state-of-the-art performance on multiple 3D spatial reasoning benchmarks. Extensive experiments and ablations demonstrate that our multi-level fusion strategy consistently enhances 3D understanding and generalizes robustly across diverse spatial reasoning tasks, establishing SpatialStack as an effective and extensible design paradigm for vision-language-geometry integration in next-generation multimodal physical AI systems.