ResearchTasksExplorative ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedCrosswords Word-level (test)CoT13.5Accuracy11Mar 9, 2026Crosswords Letter-level (test)RouteGoT23.4Accuracy11Mar 9, 2026Game of 24 (test)RTR80Accuracy11Mar 9, 2026