ResearchDatasetsZebraArenaFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsAgentic ReasoningZebraArena multi-turn Large82.24Accuracy2Agentic ReasoningZebraArena multi-turn Medium88.14Accuracy2Agentic ReasoningZebraArena multi-turn (Small)96.69Accuracy2