ResearchTasksLong-form text generationLong-form text generationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedFactScoreDHI68.1FactScore20Jul 9, 2026LongWriterQwen3-8B-WRL93.5Accuracy18May 11, 2026LongBench Write-en 1.0 (test)GPT-4o91.8Sq12Feb 26, 2026LongBenchDeepWriter-8B + RL-TCER87.3LB Score6Apr 14, 2026