ResearchTasksMulti-label Value ClassificationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedGemini-annotated N=4,000 (test)XLM-RoBERTa-large0.752Self-direction Score1Jun 11, 2026