ResearchTasksFrontier Discovery and Ambiguity DisambiguationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedNotepad fixed three-start replay-start (diagnostic slice)Reactive gpt-5-mini13.67MV(49)5May 18, 2026