ResearchBenchmarksOffline Reinforcement Learning on D4RL Adroit (8 tasks)Follow71Normalized ReturnFQL64.7666.386869.62May 7, 2026Evaluation ResultsMethodMethodLinksNormalized ReturnSuccess RateFQLSeeds=8Seeds=82026.0571—IQLSeeds=8Seeds=82026.0570—CODACSeeds=8Seeds=82026.0569—PCBFSeeds=8Seeds=82026.0569—IQNSeeds=8Seeds=82026.0566—Value FlowsSeeds=8Seeds=82026.0565—