ResearchTasksLLM-as-a-Judge CalibrationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedArena100K (test)Test oracle envelope0.227Test Risk (MSE)7Jun 2, 2026SummEval (test)Test oracle envelope0.043Test Risk (MSE)7Jun 2, 2026LLMBar (test)Test oracle envelope0.194Test Risk (MSE)7Jun 2, 2026RewardBench (test)Test oracle envelope0.022Test Risk (MSE)7Jun 2, 2026