ResearchBenchmarksOffline Reinforcement Learning on AntMaze Umaze (D4RL Normalized Score)Offline Reinforcement Learning on AntMaze Umaze (D4RL Normalized Score)Follow96.4D4RL Normalized ScoreQ-ALIGN DT63.1271.7680.489.04May 27, 2026Evaluation resultsMethodMethodLinksD4RL Normalized ScoreQ-ALIGN DT2026.0596.4QCS2026.0592.7QT2026.0590.7IQL2026.0587.5DC2026.0585TD3+BC2026.0578.6CQL2026.0574CGDT2026.0571DT2026.0565.6RVS2026.0564.4