ResearchTasksText-to-JSONFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSTAGE-EvalQwen3-4B-Thinking-250770.39PFR12Jun 19, 2026DeepJSONEvalQwen3-4B SFT + STAGE93.38Format Adherence12Jun 19, 2026