ResearchBenchmarkstask1 on puzzle 4x6Follow10,000Success RateTRL-4002,3005,0007,700Oct 26, 2025Evaluation ResultsMethodMethodLinksSuccess RateTRL2025.1010,000MC2025.109,800IQL2025.108,700TD2025.106,100SAC+BC2025.104,800FBC2025.100CRL2025.100QRL2025.100TDP2025.100COE2025.100