ResearchTasksFine-grained Score AccuracyFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMultiDialoginterpretable AI judge65.62Exact Accuracy1Mar 4, 2026Fisherinterpretable AI judge64.76Exact Accuracy1Mar 4, 2026CosyVoice 2interpretable AI judge64.5Exact Accuracy1Mar 4, 2026Oursinterpretable AI judge70.56Exact Accuracy1Mar 4, 2026