ResearchDatasetsDeep Research BenchDeep Research BenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsDeep Research EvaluationDeep Research Bench first training epoch (step 600)52.09Readability17Deep Research EvaluationDeep Research Bench (step 1100)53.81Readability16