ResearchBenchmarksOffline Reinforcement Learning on D4RL Adroit relocate (human, cloned, expert)Follow36.7Normalized ScoreReBRAC30.77232.31133.8535.389May 27, 2026Evaluation ResultsMethodMethodLinksNormalized ScoreReBRAC2026.0536.7BC2026.0536MoMa QL2026.0536IDQL2026.0535.7FQL2026.0535.7IQL2026.0535.3SRPO2026.0535.3FBRAC2026.0535.3FAWAC2026.0535IFQL2026.0534.7CAC2026.0531