ResearchDatasetspartitionedFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLatent reasoningpartitioned-k=2 query-blind (100 problems)90Accuracy6Latent reasoningpartitioned-k=2 query-known (100 problems)94Accuracy6