ResearchTasksCommonsense Fact VerificationCommonsense Fact VerificationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCREAK (test)RACO88.6Accuracy17Jun 4, 2026CREAK Contrast Set (contra)Human92.2Accuracy4Feb 26, 2026