Visual Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
66.9Accuracy (Clean)
7
Jun 4, 2026
3.213CIDEr
7
Feb 26, 2026
0.254BLEU-1
7
Feb 26, 2026
69.8Presence Precision@1
7
Feb 26, 2026
58.65VQA Accuracy
7
Feb 26, 2026
51.76F1 Score
7
Feb 26, 2026
75.9Accuracy (Single)
7
Feb 26, 2026
58.3Accuracy
7
Feb 26, 2026
58.3TextVQA Accuracy
7
Feb 26, 2026
70.8TextVQA
7
Feb 26, 2026
70.8TextVQA
7
Feb 26, 2026
70.8TextVQA
7
Feb 26, 2026
64.8Score
7
Feb 26, 2026
55.1Score
7
Feb 26, 2026
87Accuracy
7
Mar 6, 2026
77.7Accuracy
7
Mar 6, 2026
61.3Exact Match
7
Feb 26, 2026
71.5Accuracy
7
Feb 27, 2026
0.637GPT Score
7
Feb 26, 2026
89.9Human Evaluation Score
7
Feb 26, 2026
58.35Overall Accuracy
7
Feb 26, 2026
67.33Accuracy
7
Feb 26, 2026
87Accuracy
7
Feb 26, 2026
95.5Overall Accuracy
7
Feb 26, 2026
41.56Accuracy
7
Feb 27, 2026