ResearchTasksActive ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAR-Bench-DCBALAR61Exact Accuracy23May 18, 2026ARBenchOpenAI o3-mini (medium)46.7DC Score16May 8, 2026ARBench-GNTeamTR (Homogeneous)45.3Accuracy13May 18, 2026ARBench SPo4-mini43.1Accuracy13May 18, 2026AR-Bench-SPBALAR54Non-strict Semantic Equivalence10May 8, 2026