ResearchTasksImage Generation AssessmentFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMMRB2 (test)UnifiedReward-Flex69.2Accuracy8Feb 26, 2026GenAI-Bench Image (test)UnifiedReward-Flex73.4Accuracy8Feb 26, 2026