Multimodal Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
67.53Accuracy
18
Jun 2, 2026
71.56Accuracy
18
Jun 2, 2026
75.5Overall Accuracy
17
Apr 23, 2026
48.6F1 Score
16
Jun 11, 2026
69.4F1 Score
16
Jun 11, 2026
54.9F1 Score
16
Jun 11, 2026
13.8F1 Score
16
Jun 11, 2026
80VQAv2 Accuracy
15
Mar 4, 2026
65.94CP
15
Feb 26, 2026
80.8Comprehensiveness
14
Apr 21, 2026
92Comprehensiveness
14
Apr 21, 2026
99.2Accuracy
14
Mar 6, 2026
95.6Accuracy
14
Mar 6, 2026
80.4Accuracy
14
Mar 6, 2026
83.6MMBench Score
13
May 13, 2026
70.3Accuracy
13
May 6, 2026
73Accuracy
13
Apr 21, 2026
77.3Accuracy
13
Apr 21, 2026
95Retrieval F1 (BoW)
13
Apr 14, 2026
93.5Yes/No Accuracy
13
Mar 17, 2026
65.15EM
12
Feb 26, 2026
70.4Accuracy
12
Feb 26, 2026
15.9EM
11
Apr 3, 2026
38.8Exact Match (EM)
11
Apr 3, 2026
58.6EM
11
Apr 3, 2026