ResearchDatasetsOOD AverageFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsQuestion AnsweringOOD Average GPQA & MMLU-Pro83.3Accuracy9RoutingOOD Average (HotpotQA, GPQA, SVAMP)62.6Acc5