SpatialSkill is proposed, a weight-update-free framework that enables a frozen vision-language model to accumulate explicit natural-language reasoning skills from offline trajectories without modifying model parameters.
Abstract
Cross-view spatial reasoning requires a model to align different viewpoints into a coherent spatial representation, yet this ability remains challenging for vision-language models despite being natural to humans. Existing methods typically improve spatial reasoning by updating model weights, which keeps the acquired knowledge implicit and tied to a specific backbone. We propose \textit{SpatialSkill}, a weight-update-free framework that enables a frozen vision-language model to accumulate explicit natural-language reasoning skills from offline trajectories. Unlike symbolic tasks, perceptual skills cannot be reliably verified simply by executing them: a plausible spatial rule may lack visual support or require transformations that the frozen model cannot perform. SpatialSkill therefore admits candidate skills only after visual-grounding and executability checks, constrains manual evolution to prevent harmful regressions, and routes skills by spatial-reasoning category to reduce negative transfer. On CityCube, across four frozen executors, SpatialSkill yields consistent gains, and a 9B executor equipped with SpatialSkill surpasses the strongest closed-source reference in our evaluation. The skills are stored in a versioned natural-language manual, making the reasoning strategies explicit and auditable without modifying model parameters. Code at https://github.com/vindahi/SpatialSkill.
When solving complex spatial reasoning tasks, ordinary humans naturally revisit the scenes from task-relevant viewpoints and mentally reconstruct the spatial layouts from those viewpoints. But can Large Vision-Language Models (LVLMs) proactively think in the same way? We term this ability Active Spatial Mentalization (...
Yuhong Liu, Shen-Xi Wu, Jun-Yi Chi et al.· 0 citations
Large Vision-Language Models (LVLMs) have made remarkable progress across visual perception tasks, yet spatial reasoning remains a persistent weakness, especially for questions that require reasoning over visual space. Recent spatial-reasoning methods incorporate generated grounding, where models predict bounding boxes...
Rafi Ibn Sultan, Xiang-Yu Zhou, Mohammad O. S. Chowdhury et al.· arXiv.org· 0 citations
Spatial reasoning benchmarks evaluate vision-language models across diverse tasks, but task-level scores do not reveal which underlying capabilities account for success or failure. Each task requires recovering spatial evidence, representing geometry, and reasoning over it. We disentangle these capabilities by comparin...
Peng-Zhan Sun, Jun-Bin Xiao, Ramanathan Rajaraman et al.· 0 citations
Existing spatial reasoning benchmarks mainly test spatial perception: reading off relations already visible in the input. Yet real-world spatial intelligence demands predictive spatial reasoning: constructing a scene from observations, anticipating how an intervention changes it, and reasoning about the unseen outcome....
Hong-Xing Li, Jin-Yue Su, Ding-Ming Li et al.· 1 citation
Spatial intelligence requires maintaining a coherent understanding of the world as the embodied agent moves. Like humans, the agent must use its own motion to interpret changes across observations and update object locations and spatial relations accordingly. Despite spatial post-training having substantially broadened...
Shi-Chao Li, Mei-Qi Wang, Fei Su et al.· 0 citations
The results establish goal-conditioned state comparison as an explicit formulation of egocentric spatiotemporal reasoning for task-progress understanding as part of a unified framework for diagnosing and improving order-robust task-progress understanding.
Xiao-Da Yang, Can Wang, Yu-Xiang Liu et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.