ResearchTasksAverage across tasksFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAverage across tasksCLP57.2Performance @ 8k Context Length13Mar 19, 2026Long-context benchmarksL2A45.9Performance (8k Context)8Mar 19, 2026