ResearchBenchmarksOffline Reinforcement Learning on Ant Medium D4RLFollow96.4Normalized ScoreDARA14.13635.49356.8578.207Dec 2, 2025Evaluation ResultsMethodMethodLinksNormalized ScoreDARA2025.1296.4DROCO2025.1294.5OTDF2025.1292.5IGDF2025.1291.6IQL*variant=tuned/reproducedvariant=tuned/reproduced2025.1289.1CQL*variant=tuned/reproducedvariant=tuned/reproduced2025.1257.3BOSA2025.1236.1DARA2025.1232.9OTDF2025.1232.3BOSA2025.1230.1DVDF-IGDF2025.1228.1DVDF-OTDF2025.1226.4IGDF2025.1225.6IQL2025.1217.3