ResearchTasksConsistency EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedDecisionBenchDebate-AHP0.29CR Mean9Feb 26, 2026Diagnostic (Avg. YouCook2, COIN, CrossTask) (test)CAST76.92State Accuracy8Mar 10, 2026RealUnify GEUBagel + SFT0.32MC Score4May 19, 2026Unified-BenchBagel + Unicot90.26CLIP Score4May 19, 2026Consistency Evaluation Dataset (N=720) 1.0 (test)——Primary metric0Feb 26, 2026