ResearchBenchmarksOffline Reinforcement Learning on D4RL MuJoCo Locomotion TotalFollow91.5Normalized ScoreISEP-FM50.31661.00871.782.392May 18, 2026Evaluation ResultsMethodMethodLinksNormalized ScoreISEP-FMFlow Matching=trueFlow Matching=true2026.0591.5QIPO2026.0588.1DQL2026.0588QGPO2026.0586.6ISEPFlow Matching=falseFlow Matching=false2026.0584IDQL2026.0579.1IQL2026.0577CQL2026.0574.6BC2026.0551.9