Vision-Language Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
74MVista Score
28
Feb 26, 2026
79.8MathVista Acc
25
Feb 26, 2026
87.7Overall Accuracy
22
Jun 5, 2026
66.67RAE
22
Jun 5, 2026
44.1Average Score
16
May 13, 2026
74.7Accuracy
14
Jun 4, 2026
54.9BLEU-1
13
Mar 24, 2026
43.3BLEU-1
13
Mar 24, 2026
79.8Barrier
13
Feb 26, 2026
86.16Accuracy
12
Feb 26, 2026
69.38Accuracy
10
May 19, 2026
77.59Score
10
Feb 26, 2026
59.88Simple Acc
9
Feb 26, 2026
34.6Accuracy
7
May 28, 2026
62.75Simple Accuracy
7
Feb 26, 2026
66.02Simple Accuracy
7
Feb 26, 2026
78.98Simple Accuracy
7
Feb 26, 2026
53.31Simple Accuracy
7
Feb 26, 2026
88.4Simple Accuracy
7
Feb 26, 2026
31.58Road Damage (Perturbed) Performance
4
Jul 8, 2026
4.67Comprehensiveness of Findings
4
Apr 30, 2026
85.6General Score (SSRBench)
4
Mar 31, 2026
67BERT F1 Score
4
Mar 17, 2026
30.5Text Score
4
Feb 26, 2026
46.2Average Accuracy
3
Mar 16, 2026