ResearchTasksMany-to-one Image Generation ((T+A) → I)FollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedTriplet (text, image, audio)FlowBind25.57CLIP Score (T→I)3Feb 26, 2026