ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo halfcheetah-mediumFollow48.4Normalized ScoreTD3+BC39.66441.93244.246.468Jun 12, 2026Evaluation ResultsMethodMethodLinksNormalized ScoreTD3+BCSeeds=5Seeds=52026.0648.4IQLSeeds=5Seeds=52026.0647.4DRIVESeeds=5Seeds=52026.0646.7CQLSeeds=5Seeds=52026.0644.4PDiTSeeds=5Seeds=52026.0642.8BEARSeeds=5Seeds=52026.0641.7BCSeeds=5Seeds=52026.0640.5DTSeeds=5Seeds=52026.0640