ResearchTasksTextual understandingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedArena-HardUni-DPO67.1Win Rate17May 26, 2026AlpacaEval 2Uni-DPO54.7Length Constraint (LC)17May 26, 2026SedarEvalSimPO57.7Overall Accuracy15May 26, 2026IFEvalUni-DPO71.2Strict Accuracy15May 26, 2026