ResearchTasksPost-RL EvaluationPost-RL EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedHealthBench HardMRRG32.1Accuracy5Jul 3, 2026BiGGen BenchMRRG63.7Accuracy5Jul 3, 2026