ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo halfcheetah-expertFollow93Normalized ScoreDRIVE51.81662.50873.283.892Jun 12, 2026Evaluation ResultsMethodMethodLinksNormalized ScoreDRIVESeeds=5Seeds=52026.0693DTSeeds=5Seeds=52026.0691.7TD3+BCSeeds=5Seeds=52026.0690.7IQLSeeds=5Seeds=52026.0686.7BCSeeds=5Seeds=52026.0686.2PDiTSeeds=5Seeds=52026.0673CQLSeeds=5Seeds=52026.0662.4BEARSeeds=5Seeds=52026.0653.4