ResearchDatasetsLongBench and Needle-In-A-HaystackFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLong-context reasoningLongBench and Needle-In-A-Haystack (NIAH) (test)29.33MultiFieldQA Score5