Visual Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
36.64Accuracy
4
Feb 26, 2026
96.11Scene Recognition
4
Feb 26, 2026
97.7Accuracy (size 4)
4
Feb 26, 2026
72.53Accuracy
4
Feb 26, 2026
44.44Accuracy
4
Feb 26, 2026
57.66Open Accuracy
3
Jun 18, 2026
87.4Accuracy
3
Jun 17, 2026
63.08Accuracy
3
Jun 11, 2026
70.08Accuracy
3
Jun 11, 2026
59.95Accuracy (Typical)
3
Jun 11, 2026
191.1C Score
3
Jun 5, 2026
92.1Accuracy
3
May 20, 2026
85.1Adv Accuracy
3
May 20, 2026
52.23Accuracy
3
May 19, 2026
14.55STVQA Score
3
Apr 14, 2026
50.8EM
3
Mar 31, 2026
52.5Exact Match
3
Mar 31, 2026
52.1Exact Match
3
Mar 31, 2026
54.3EM
3
Mar 31, 2026
49.6Exact Match (EM)
3
Mar 31, 2026
0.952GPT Judge Detailed Score
3
Mar 27, 2026
58.5Accuracy
3
Mar 24, 2026
72.2Accuracy
3
Mar 11, 2026
22.02Avg (N)
3
Mar 11, 2026
78.53Accuracy
3
Mar 4, 2026