ResearchTasksLLM fine-tuning data selectionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedDATE-LM MMLU, GSM8K, BBH official (test)BipCov (ref-aligned)61.87MMLU Accuracy3Jun 1, 2026