Audio-Visual Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
13.8ASR Rate
7
Feb 26, 2026
96.03ASR Accuracy
7
Feb 26, 2026
95.6AVQA Clean Accuracy
7
Feb 26, 2026
84.3Overall Accuracy
7
Feb 26, 2026
40.4Error Rate (ER)
6
May 26, 2026
79.5Original Score
6
Mar 10, 2026
152.9CIDEr
6
Feb 26, 2026
54.8Accuracy
6
Feb 26, 2026
69.8Accuracy
6
Feb 26, 2026
79.44Accuracy (Audio)
5
Mar 5, 2026
93.8Accuracy
5
Feb 26, 2026
62.2CIDEr
5
Feb 26, 2026
83.82Accuracy (Audio, Count)
4
Feb 26, 2026
58.86Total Accuracy
4
Feb 26, 2026
57.5Average Score
3
Jul 8, 2026
40.2Average Score
3
Jul 8, 2026
56.53M.J. Score
3
Feb 26, 2026