ResearchTasksKnowledge QAFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedCDDMBenchQwen-VL-Chat-AG* (7B)88.5QA Accuracy31Apr 28, 2026RoleScape-20UniCharacter77Knowledge QA Accuracy3May 12, 2026GPQAGLM-4.7-FP859.6Accuracy3May 7, 2026