SpatialClaw: Training-Free Code-as-Action Interface Boosts VLM Agentic Spatial Reasoning

CMHungSteven · x · 2026-09-29

A new arXiv paper from an NVIDIA internship team studies how the action interface shapes VLM spatial reasoning agents:

Related event: NVIDIA's SpatialClaw: Code as Action Interface Boosts VLM Spatial Reasoning Training-Free, Accepted at NeurIPS(5 posts)→

Original post →

More from Research

Research channel →