ResearchTasksMetric-Human Agreement EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedHuman-agreement n=150 (test)Spider2 M60.502Delta-Kappa (Hybrid-EX vs Baseline)7Jun 2, 2026