ResearchDatasetsRuleArenaFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsRule-governed ReasoningRuleArena NBA44.2Strict Accuracy14Rule-following ReasoningRuleArena NBA (test)74.2Accuracy6Rule-based ReasoningRulearena NBA67Accuracy3Rule-based ReasoningRulearena Tax50Accuracy3