ResearchTasksLong-context Information RetrievalLong-context Information RetrievalFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedNeedle-In-a-Haystack Verbatim prompt (test)Vanilla0.996Accuracy (Depth 0%)7Feb 26, 2026