PlanarBench: Evaluating LLM Spatial Reasoning via Planar Graph Drawing 事件

PlanarBench: Evaluating LLM Spatial Reasoning via Planar Graph Drawing arXiv:2606.02010v1 Announce Type: new Abstract: PlanarBench tests whether LLMs can draw planar graphs as ASCII art given only an edge list -- a spatial reasoning task that resists memorization because edge order, edge orientation, and node labels are all permutable. We evaluate 91 models on the 199 simplest non-isomorphic connected planar graphs (2 - 7 vertices). Edge count is the dominant difficulty predictor ($r = -0.8