ResearchTasksAgent Task and Scenario CompletionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAppWorld (dev)Trajectory-Informed Memory Generation89.5Task Goal Completion2Mar 12, 2026AppWorld (train)Trajectory-Informed Memory Generation91.1Task Goal Completion2Mar 12, 2026