ResearchTasksReasoning failure prediction and recoveryReasoning failure prediction and recoveryFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCRUXEval L2thought-tree-based classifier77Accuracy4Apr 21, 2026SAFIM L3thought-tree-based classifier81Accuracy2Apr 21, 2026SAFIM L1thought-tree-based classifier79Accuracy2Apr 21, 2026CRUXEval (L3)thought-tree-based classifier74Accuracy2Apr 21, 2026CRUXEval L1thought-tree-based classifier89Accuracy2Apr 21, 2026