ResearchDatasetsMAVINSetMAVINSetFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMulti-shot Audio-Visual GenerationMAVINSet high-fidelity benchmark 1K-sample (test)231.6FVD11Multi-shot Audio-Visual GenerationMAVINSet subjective 20 samples36.8AVQ11Text-to-Audio-Video GenerationMAVINSet 1.0 (test)231.6FVD3