ResearchBenchmarksOffline Reinforcement Learning on Walker2D Medium-Expert (Noise 12)Follow93.5Normalized ReturnI-TAP29.935246.437662.9479.4424Feb 21, 2026Evaluation ResultsMethodMethodLinksNormalized ReturnI-TAPModel Category=Model-B...Model Category=Model-Based2026.0293.5L-MAPModel Category=Model-B...Model Category=Model-Based2026.0292.75TAPModel Category=Model-B...Model Category=Model-Based2026.0291.09CQLModel Category=Model-FreeModel Category=Model-Free2026.0283.18IQLModel Category=Model-FreeModel Category=Model-Free2026.0268.61DTModel Category=Model-FreeModel Category=Model-Free2026.0252.421R2RModel Category=Model-B...Model Category=Model-Based2026.0232.38