ResearchTasksMulti-hop reasoning and question answeringFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedPopQAAgentConductor50.3PopQA Score2Feb 26, 2026HLEAgentConductor22.6HLE Average2Feb 26, 2026GAIAAgentConductor72GAIA L1 Acc2Feb 26, 2026