ResearchBenchmarksOffline Goal-Conditioned Reinforcement Learning on puzzle 4x4 play v0Offline Goal-Conditioned Reinforcement Learning on puzzle 4x4 play v0Follow88Average Binary Success RateARLe-3.5220.244467.76May 21, 2026Evaluation resultsMethodMethodLinksAverage Binary Success RateARLeSize=1M, Dim.=83Size=1M, Dim.=832026.0588ARLiSize=1M, Dim.=83Size=1M, Dim.=832026.0578HIQL1vrSize=1M, Dim.=83Size=1M, Dim.=832026.0549IQLSize=1M, Dim.=83Size=1M, Dim.=832026.0530HIQL2vrSize=1M, Dim.=83Size=1M, Dim.=832026.0517HIQL2vSize=1M, Dim.=83Size=1M, Dim.=832026.050