Visual Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
72.9F1 Score
5
Mar 4, 2026
78.6EM
5
Mar 4, 2026
68Accuracy (clean)
5
Feb 26, 2026
59.6Accuracy (Clean)
5
Feb 26, 2026
73.2Accuracy (clean)
5
Feb 26, 2026
60.5F1 (Q1)
5
Feb 26, 2026
79.65Exact Match
5
Feb 26, 2026
63Accuracy
5
Feb 26, 2026
84.11Accuracy
5
Feb 26, 2026
44.7Accuracy
5
Feb 26, 2026
70.85ACC
5
Feb 26, 2026
71.59Accuracy
5
Feb 26, 2026
69.93Accuracy
5
Feb 26, 2026
73.8Accuracy
5
Feb 26, 2026
73.25Accuracy
5
Feb 26, 2026
60.9Recall
5
Feb 26, 2026
84.3Accuracy
5
Feb 26, 2026
32.3RQ
5
Feb 26, 2026
86.4Accuracy
5
Feb 26, 2026
48.74VQA Score
5
Feb 26, 2026
68.63VQA Score
5
Feb 26, 2026
70.5Overall Accuracy
5
Feb 26, 2026
0.23FTL (s)
5
Feb 26, 2026
1,511Perception Score
5
Feb 26, 2026
58.8Top-1 Acc
5
Feb 26, 2026