ResearchDatasetsPrinciple-based evaluationFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsOverallPrinciple-based evaluation dataset8.41Average12SteeringPrinciple-based evaluation dataset8.68G Score12JudgmentPrinciple-based evaluation dataset8.45G Score12