ResearchTasksLong-document understandingLong-document understandingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedLong-document Understanding BenchmarkHMARS83Synonymy Score12Jun 30, 2026NarrativeQAQwen3-235B-A22B-Thinking88.5Score (avg@3)3May 22, 2026