ResearchBenchmarkstask3 on puzzle 4x6Follow6,700Success RateTRL-2681,5413,3505,159Oct 26, 2025Evaluation ResultsMethodMethodLinksSuccess RateTRL2025.106,700MC2025.103,400FBC2025.100IQL2025.100CRL2025.100SAC+BC2025.100QRL2025.100TDP2025.100COE2025.100TD2025.100