ResearchTasksAutomatic Metric Correlation with Human Factual Correctness JudgmentsFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedYouCook3-Fact and CraftBench-Fact (User Study)MultiFactScore0.534Pearson r8Apr 29, 2026