SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning Paper • 2606.13673 • Published Jun 11 • 111
SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning Paper • 2606.13673 • Published Jun 11 • 111
FineBench: Benchmarking and Enhancing Vision-Language Models for Fine-grained Human Activity Understanding Paper • 2605.19846 • Published May 20 • 3
Physics in 2-Steps: Locking Motion Priors Before Visual Refinement Erases Them Paper • 2606.06361 • Published Jun 4 • 16