ResearchBenchmarksOffline Reinforcement Learning on Hopper Medium Noise 2.5Follow72.47Normalized ReturnI-TAP42.486850.270958.05565.8391Feb 21, 2026Evaluation ResultsMethodMethodLinksNormalized ReturnI-TAPModel Category=Model-B...Model Category=Model-Based2026.0272.471R2RModel Category=Model-B...Model Category=Model-Based2026.0265.24L-MAPModel Category=Model-B...Model Category=Model-Based2026.0263.21DTModel Category=Model-FreeModel Category=Model-Free2026.0260.97IQLModel Category=Model-FreeModel Category=Model-Free2026.0256CQLModel Category=Model-FreeModel Category=Model-Free2026.0249.92TAPModel Category=Model-B...Model Category=Model-Based2026.0243.64