ResearchTasksPointwise Reasoning EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedDeepfakeJudge Meta-HumanDeepfakeJudge-7B0.5RMSE12Feb 26, 2026DeepfakeJudge MetaDeepfakeJudge-7B0.61RMSE12Feb 26, 2026