ResearchTasksText-to-JSONText-to-JSONFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedSTAGE-EvalQwen3-4B-Thinking-250770.39PFR12Jun 19, 2026DeepJSONEvalQwen3-4B SFT + STAGE93.38Format Adherence12Jun 19, 2026