ResearchTasksSafe Continual Reinforcement LearningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedDamaged HalfCheetah VelocityPPO2,078.8Final Reward8Apr 23, 2026Safe Continual WorldPPO89Success Rate7Apr 23, 2026Damaged Ant VelocityCF-EWC3,739.7Final Reward7Apr 23, 2026