ResearchTasksAudio-Visual Scene-Aware DialogAudio-Visual Scene-Aware DialogFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAVSD (test)VX2TEXT1.605CIDEr11Feb 26, 2026AVSD (val)Combined Loss59.48ASR (%)7Feb 26, 2026AVSDCAT (ours)145.5CIDEr6Mar 27, 2026