ResearchBenchmarkstask2 on puzzle 4x6Follow66Success RateTRL-2.6415.183350.82Oct 26, 2025Evaluation ResultsMethodMethodLinksSuccess RateTRL2025.1066MC2025.1046SAC+BC2025.105TD2025.102FBC2025.100IQL2025.100CRL2025.100QRL2025.100TDP2025.100COE2025.100