SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild? Paper • 2602.03916 • Published 5 days ago • 11