ResearchBenchmarksReinforcement Learning on Antmaze umaze offline-to-onlineReinforcement Learning on Antmaze umaze offline-to-onlineFollow99.4ScoreRLPD89.72892.23994.7597.261May 14, 2026Evaluation resultsMethodMethodLinksScoreRLPDEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0599.4CQLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0599CPQLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0598.2PEXEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0595.2Cal-QLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0590.1