ResearchTasksLong-context language evaluationLong-context language evaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedLongBench v1 (test)LOOKAHEADKV19.6NrtQA Score31Mar 12, 2026LongBenchBaseline48.43SQA19May 25, 2026