ResearchDatasetsK&KK&KFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsLogical ReasoningK&K tasks2.13Perplexity (3ppl)32Data Contamination DetectionK&K70F1 Score16Logic ReasoningK&K25Winrate1