ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo halfcheetah-medium-replayFollow46.2Normalized ScoreCQL34.55237.57640.643.624Jun 12, 2026Evaluation ResultsMethodMethodLinksNormalized ScoreCQLSeeds=5Seeds=52026.0646.2TD3+BCSeeds=5Seeds=52026.0644.6IQLSeeds=5Seeds=52026.0644.2DRIVESeeds=5Seeds=52026.0643.7PDiTSeeds=5Seeds=52026.0640.8BEARSeeds=5Seeds=52026.0638.6BCSeeds=5Seeds=52026.0635.8DTSeeds=5Seeds=52026.0635