ResearchTasksMulti-turn agent taskMulti-turn agent taskFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedSearch-QAGRPO50.49Success Rate31Jul 1, 2026ACEBench multi-turn (test)H-EPM76.5Process Accuracy31Apr 29, 2026ALFWorldATOD85.16Success Rate23Jun 29, 2026