ResearchTasksArithmetical puzzle solvingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedArithmetical Puzzle OOD-1000 1.0GPT40.1pass@13Feb 26, 2026Arithmetical Puzzle 1.0 (OOD-F)Deepseek-Coder-33B0.1Pass@13Feb 26, 2026Arithmetical Puzzle 1.0 (OOD-N)GPT40.001Pass@13Feb 26, 2026Arithmetical Puzzle 1.0 (ID)GPT40.2Pass@13Feb 26, 2026