ResearchBenchmarksOffline Reinforcement Learning on Walker2D Medium-Replay Noise 7Follow74.04Normalized ReturnI-TAP38.88848.01457.1466.266Feb 21, 2026Evaluation ResultsMethodMethodLinksNormalized ReturnI-TAPModel Category=Model-B...Model Category=Model-Based2026.0274.04L-MAPModel Category=Model-B...Model Category=Model-Based2026.0269.21R2RModel Category=Model-B...Model Category=Model-Based2026.0252.23DTModel Category=Model-FreeModel Category=Model-Free2026.0249.51TAPModel Category=Model-B...Model Category=Model-Based2026.0243.49IQLModel Category=Model-FreeModel Category=Model-Free2026.0240.77CQLModel Category=Model-FreeModel Category=Model-Free2026.0240.24