ResearchTasksMulti-agent DebateFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMixed regimeFlat72.4Accuracy15Jun 18, 2026Social regimeSituational leadership controller51.3Accuracy15Jun 18, 2026Closed regimeMAD87.7Accuracy15Jun 18, 2026AlphaNLIR0-vote93.7Accuracy15Jun 18, 2026MMLU-Pro DebateConcise65.7Success Rate (SR)10Apr 9, 2026