ResearchTasksReward Hacking Onset LocalizationReward Hacking Onset LocalizationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCHERRL VerInstruct and HealthBench runsCC-Qwen490VerInstruct SP8Jun 4, 2026