ResearchTasksLong-Form Document-Style Memory ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMuSiQueREAL30.34Exact Match (EM)24Jun 11, 20262WikiMultihopQAREAL45.65Exact Match (EM)24Jun 11, 2026HotpotQAREAL36.4EM24Jun 11, 2026