Visual Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
80.2In-domain Accuracy
6
Feb 26, 2026
85.96Accuracy
6
Feb 27, 2026
74.37Overall Accuracy (MC)
6
Feb 26, 2026
44.48Accuracy
6
Feb 26, 2026
0.679AUROC
6
Feb 26, 2026
0.354ECE
6
Feb 26, 2026
76.5Location Accuracy
5
Jun 16, 2026
61.03Overall Accuracy
5
Jun 5, 2026
74Accuracy (NLVR)
5
Jun 2, 2026
63.2Accuracy
5
May 29, 2026
73.7F1 Score
5
May 27, 2026
3.27Gain Score
5
May 21, 2026
0.441BLEU
5
Apr 24, 2026
0.474BLEU
5
Apr 24, 2026
60.23VQA Score
5
Apr 20, 2026
56.25Accuracy (Cambodia)
5
Apr 14, 2026
27.96GRP Score
5
Apr 14, 2026
0Refusal Rate (Exact)
5
Apr 13, 2026
63.93BLEU-1
5
Mar 27, 2026
68Score
5
Mar 23, 2026
76.39Accuracy
5
Mar 13, 2026
75.86Accuracy (ACC)
5
Mar 13, 2026
46.06Accuracy
5
Mar 13, 2026
49.74Accuracy
5
Mar 13, 2026
61.2Accuracy
5
Mar 4, 2026