ResearchBenchmarksLocomotion on D4RL walker2d-medium-replayFollow3,491.2Mean ReturnCOOPO436.0961,229.2482,022.42,815.552May 18, 2026Evaluation ResultsMethodMethodLinksMean ReturnCOOPO2026.053,491.2TD3+BC2026.053,252.7CQL2026.053,073.5Uni-O42026.053,072.3PEX2026.052,503.7Cal-QL2026.052,271.8IQL2026.052,076.4Onestep RL2026.051,972.6ODT2026.051,380.2AWAC2026.051,075.8BC2026.051,037.3Off2ON2026.05553.6