ResearchTasksSemantic consistency evaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedTIFAVisualPrompter93.8Avg Answering Accuracy20Mar 9, 2026DSGVisualPrompter84.3Average Answering Accuracy20Mar 9, 2026SMPL mesh (test)ISS7.8mIoU (t=0.00)7Feb 26, 2026Semantic-stress diagnostic dataset (evaluation subset)Gemma3-4B0.009Mean ∆lex5Jun 9, 2026