ResearchBenchmarksGoal-oriented Robotic Manipulation on FetchPush 10-p v2Follow88.1Normalized ReturnTD3BC67.61272.93178.2583.569Feb 11, 2026Evaluation ResultsMethodMethodLinksNormalized ReturnTD3BC2026.0288.1IQL2026.0284.1Flex-f-Q2026.0281.3CQL2026.0276.1Opti-DICE2026.0269.5Flex-f-DICE2026.0268.4