ResearchDatasetsMulti-hopFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMulti-hop tool-use evaluationMulti-hop 450 tasks (test)89Accuracy18Information RetrievalMulti-hop58.16NDCG@1012Multi-hop RetrievalMulti-hop 4 datasets aggregate (test)58.5NDCG@108Multi-hop reasoningMulti-hop 2-hop N=50079Accuracy2