ResearchBenchmarksOffline Reinforcement Learning on D4RL ant medium-expertFollow120.3Normalized ScoreDROCO7.9837.1466.395.46Dec 2, 2025Evaluation ResultsMethodMethodLinksNormalized ScoreDROCO2025.12120.3IGDF2025.12116.8IQL*variant=tuned/reproducedvariant=tuned/reproduced2025.12113.1DARA2025.12111.9CQL*variant=tuned/reproducedvariant=tuned/reproduced2025.12107.3OTDF2025.12107.3BOSA2025.12100.5DVDF-OTDF2025.1223.2DVDF-IGDF2025.1221.1OTDF2025.1219.1IGDF2025.1217.5BOSA2025.1215.8IQL2025.1215.4DARA2025.1212.3