ResearchDatasetsSyntheticsFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMetric alignment with human overall quality judgementsSynthetics BSL0.83Pearson r20Metric alignment with human understandability judgementsSynthetics BSL0.85Pearson r20