ResearchDatasetsTrustworthinessFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsText ClassificationTrustworthiness Plain-text OOD68.44Accuracy4Trustworthiness EvaluationTrustworthiness Average (human evaluation)0.88Control Win Rate2Text ClassificationTrustworthiness Plain-text OOD (test)—Accuracy0