ResearchBenchmarksGrid-based maze navigation on MazeNamo 12x12 MediumFollow100Success RatePure FD89.692.39597.7Apr 29, 2026Evaluation ResultsMethodMethodLinksSuccess RateSuccess Rate DeltaTime (s)Path LengthPure FDConfig=Pure FDConfig=Pure FD2026.04100—6.1921.1LLM-FlaxRules=LLM-FlaxRules=LLM-Flax2026.0498—2.15122.3LLM-FlaxConfig=LLM-FlaxConfig=LLM-Flax2026.0498—2.15122.3ManualRules=ManualRules=Manual2026.049621.91422.1ManualConfig=ManualConfig=Manual2026.0496—1.91422.1PLOIConfig=PLOIConfig=PLOI2026.0490—1.3221.8