ResearchBenchmarksOffline Reinforcement Learning on OGBench cube-double-singletaskOffline Reinforcement Learning on OGBench cube-double-singletaskFollow45Normalized ScoreGTP4.4414.9725.536.03Oct 13, 2025Evaluation resultsMethodMethodLinksNormalized ScoreGTPobservation_type=state...observation_type=state-based, averaging=5 tasks2025.1045FQLobservation_type=state...observation_type=state-based, averaging=5 tasks2025.1029CACobservation_type=state...observation_type=state-based, averaging=5 tasks2025.106