ResearchBenchmarksOffline Reinforcement Learning on D4RL Adroit Pen Human Cloned ExpertFollow119.3Normalized ScoreReBRAC72.584.6596.8108.95May 27, 2026Evaluation ResultsMethodMethodLinksNormalized ScoreReBRAC2026.05119.3MoMa QL2026.05108.3IFQL2026.0596.7IQL2026.0596.3IDQL2026.0593.3FQL2026.0589.7SRPO2026.0588FBRAC2026.0587.7FAWAC2026.0582.3BC2026.0577.7CAC2026.0574.3