Visual Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
91.6Accuracy
8
Jun 23, 2026
85.9Accuracy
8
Jun 12, 2026
71.03Average Score
8
Jun 12, 2026
41.48Accuracy
8
Jun 11, 2026
78.51Accuracy
8
Jun 11, 2026
51.2Accuracy
8
Jun 4, 2026
51.66Accuracy
8
Jun 4, 2026
58.37Accuracy
8
Jun 4, 2026
75.62Accuracy
8
Jun 4, 2026
63.5Judge Accuracy
8
Jun 2, 2026
39.5Object Localization & Counting Score
8
May 20, 2026
82.2Basic Judging Score
8
May 20, 2026
4.2Average Score
8
May 20, 2026
3.82Average Score
8
May 20, 2026
79.05Accuracy
8
May 19, 2026
7.57Glass Count Score
8
May 11, 2026
42.8VQA Accuracy
8
Apr 27, 2026
70.3BEM
8
Apr 27, 2026
50Accuracy
8
Apr 27, 2026
2.09Average Rank (SEA)
8
Apr 14, 2026
-23.78BD-Rate
8
Apr 2, 2026
-53.47BD-Rate
8
Apr 2, 2026
-29.3BD-Rate
8
Apr 2, 2026
67.89Accuracy
8
Mar 20, 2026
54.2ASR (LLaVA*)
8
Mar 11, 2026