ResearchTasksInstruction Following and Helpfulness EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAlpacaEval 2.0QWEN2-INSTRUCT w/ SSO_DPO49.4Win Rate58Feb 26, 2026