ResearchBenchmarksOffline Reinforcement Learning on AntMaze Ultra-DiverseOffline Reinforcement Learning on AntMaze Ultra-DiverseFollow5,460Avg Normalized ScoreGCPC-218.41,255.82,7304,204.2Jul 7, 2023Oct 23, 2023Feb 8, 2024May 26, 2024Sep 11, 2024Dec 28, 2024Apr 16, 2025Evaluation resultsMethodMethodLinksAvg Normalized ScoreGCPC2023.075,460GCIQL2023.072,840RvS-G2023.072,640TAP2023.072,600DWSL2023.072,500IQL2023.071,780CQL2023.07960LEQ2025.0455.8VIPO-LEQ2025.0426.5IQL-TD-MPC2025.043.6BC2023.070DT2023.070WGCSL2023.070MOBILE2025.040