{"ID":2826389,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.19683","arxiv_id":"2512.19683","title":"From Indoor to Open World: Revealing the Spatial Reasoning Gap in MLLMs","abstract":"While Multimodal Large Language Models (MLLMs) have achieved impressive performance on semantic tasks, their spatial intelligence--crucial for robust and grounded AI systems--remains underdeveloped. Existing benchmarks fall short of diagnosing this limitation: they either focus on overly simplified qualitative reasoning or rely on domain-specific indoor data, constrained by the lack of outdoor datasets with verifiable metric ground truth. To bridge this gap, we introduce a large-scale benchmark built from pedestrian-perspective videos captured with synchronized stereo cameras, LiDAR, and IMU/GPS sensors. This dataset provides metrically precise 3D information, enabling the automatic generation of spatial reasoning questions that span a hierarchical spectrum--from qualitative relational reasoning to quantitative metric and kinematic understanding. Evaluations reveal that the performance gains observed in structured indoor benchmarks vanish in open-world settings. Further analysis using synthetic abnormal scenes and blinding tests confirms that current MLLMs depend heavily on linguistic priors instead of grounded visual reasoning. Our benchmark thus provides a principled platform for diagnosing these limitations and advancing physically grounded spatial intelligence.","short_abstract":"While Multimodal Large Language Models (MLLMs) have achieved impressive performance on semantic tasks, their spatial intelligence--crucial for robust and grounded AI systems--remains underdeveloped. Existing benchmarks fall short of diagnosing this limitation: they either focus on overly simplified qualitative reasonin...","url_abs":"https://arxiv.org/abs/2512.19683","url_pdf":"https://arxiv.org/pdf/2512.19683v2","authors":"[\"Mingrui Wu\",\"Zhaozhi Wang\",\"Fangjinhua Wang\",\"Jiaolong Yang\",\"Marc Pollefeys\",\"Tong Zhang\"]","published":"2025-12-22T18:58:12Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}