ResearchBenchmarksReinforcement Learning on Antmaze umaze-diverse offline-to-onlineReinforcement Learning on Antmaze umaze-diverse offline-to-onlineFollow99.2ScoreRLPD32.22449.6126784.388May 14, 2026Evaluation resultsMethodMethodLinksScoreRLPDEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0599.2CPQLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0590.4CQLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0576.9Cal-QLEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0575.2PEXEvaluation Setting=off...Evaluation Setting=offline-to-online2026.0534.8