ResearchTasksClarification GenerationClarification GenerationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedDeepResearch Bench online interactive settingsIntentRL36.44Intent Precision6Feb 26, 2026DeepResearch Bench offline (test)IntentRL2.43Quality Score4Feb 26, 2026