SATURN: 多视角定位的符号化空间推理
SATURN: Symbolic Spatial Reasoning for Multi-Perspective Grounding
- Michigan State University（密歇根州立大学）
- Lambda Labs（Lambda实验室）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出SATURN框架，通过神经符号方法实现视角感知的组合空间推理，结合3D场景重建、软空间谓词和Python符号执行器，在3D FORCE和MindCube基准上显著优于强基线。
AI中文摘要：
视觉语言模型（VLM）在空间推理需要组合依赖于参考框架的关系时仍然不可靠。现有的神经符号方法使推理更加明确，但通常依赖于脆弱的几何过程和对噪声感知的硬决策。我们提出SATURN，一个用于视角感知的组合空间推理的神经符号框架。SATURN重建近似3D场景，推导软视角感知空间谓词，并通过无训练的Python符号执行器组合它们，通过多跳推理分离感知与推理，同时保留不确定性。我们还引入了3D FORCE，一个诊断基准，控制空间排列定位（SAG）和指代表达定位（REF）中的推理深度、视角和视角组合。在3D FORCE上，VLM和空间训练模型随着深度和视角复杂性的增加而急剧下降，而SATURN保持稳定并优于强基线。在真实世界的MindCube基准上，SATURN实现了78.57%的整体准确率，比最强基线高出14个百分点。
英文摘要：
Vision-Language Models (VLMs) remain unreliable when spatial reasoning requires composing relations whose meanings depend on frames of reference. Existing tool-augmented spatial reasoning methods make reasoning more explicit, but often rely on low-level geometric procedures and hard binary decisions over noisy perception. We propose SATURN, a neuro-symbolic framework for perspective-aware compositional spatial reasoning. SATURN reconstructs an approximate 3D scene, derives soft perspective-aware spatial predicates, and composes them with a training-free Pythonic symbolic executor, separating perception from reasoning while preserving uncertainty through multi-hop inference. We also introduce 3D FORCE, a diagnostic benchmark that controls reasoning depth, view, and perspective composition for spatial arrangement grounding (SAG) and referring expression grounding (REF). On 3D FORCE, VLMs and spatially trained models degrade sharply as depth and perspective complexity increase, whereas SATURN degrades the least and outperforms every baseline at each depth. On the real-world MindCube benchmark, SATURN achieves \(78.06\%\) overall accuracy, outperforming the strongest baseline by \(14\) percentage points.
