ResearchTasksLong-horizon procedural planningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedEgoPlan-Bench AllPlanAgent + Mem.58.72Success Rate13Mar 10, 2026EgoPlan-Bench Out-of-DomainGPT-5.154.37Success Rate9Mar 10, 2026EgoPlan-Bench In-DomainPlanAgent + Mem.62.46Success Rate9Mar 10, 2026