AI RESEARCH
STAR: Mitigating Cascading Errors in Spatial Reasoning via Turn-point Alignment and Segment-level DPO
arXiv CS.CV
•
ArXi:2604.00558v1 Announce Type: new Structured spatial navigation is a core benchmark for Large Language Models (LLMs) spatial reasoning. Existing paradigms like Visualization-of-Thought (VoT) are prone to cascading errors in complex topologies. To solve this, we propose STAR, a two-stage framework grounded on topological anchors, and