Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Visual Question Answering on V*Bench

98.95Accuracy

Human

18.911639.690860.4781.2492Dec 21, 2023May 17, 2024Oct 12, 2024Mar 10, 2025Aug 5, 2025Dec 31, 2025May 29, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2023.12
98.9598.26100
2026.02
92.194.888.1
2026.02
91.193.986.8
2026.02
90.693.985.5
2025.06
90.58--
2026.02
86.990.481.5
2025.10
86.9--
2026.02
86.488.785.5
2025.10
85--
2026.02
84.386.181.6
84.385.282.9
2026.02
83.285.280.3
2025.10
81.7--
2026.02
81.28082.9
80.181.777.6
2026.02
79.182.673.7
79.06--
2026.02
788075
2026.02
7879.176.3
2026.02
76.478.273.6
2023.12
75.3974.7876.31
2026.02
74.978.368.7
2025.10
74.4--
2025.10
69.6--
2025.10
68.1--
2025.10
67--
2025.10
64.9--
2026.02
62.365.257.9
2026.02
60.763.556.6
2023.12
54.9751.360.52
2023.12
48.6843.4756.57
2023.12
48.1640.8659.21
2026.04
46.6--
2026.04
46--
2026.04
46--
2026.04
45.5--
2026.04
45.5--
2026.04
44.5--
2026.04
44.5--
2026.04
44.5--
2026.04
43.9--
2026.04
43.9--
2026.04
43.4--
2026.04
43.4--
2026.04
43.4--
2026.04
42.9--
2026.04
42.9--
2026.04
42.9--
2026.04
42.4--
2026.04
42.4--
2026.04
42.4--
2026.04
41.8--
2023.12
41.3634.7851.31
2023.12
41.3631.356.57
2026.04
41.3--
2023.12
38.7426.9556.57
2026.04
38.7--
2023.12
38.2230.4350
2026.04
38.2--
2026.04
38.2--
2026.04
37.7--
2026.04
37.7--
2023.12
37.6926.9553.94
2023.12
37.6930.4348.68
2023.12
37.1731.346.05
2026.04
37.1--
2026.04
36.6--
2026.04
36.1--
2026.04
36.1--
2026.04
36.1--
2026.04
36.1--
2023.12
35.9926.7350
2026.04
35.6--
2026.04
35.6--
2026.04
35.6--
2023.12
35.5923.4753.94
2026.04
35--
2026.04
35--
2026.04
35--
2026.04
34.5--
2026.04
34.5--
2026.04
34.5--
2023.12
34.0225.2147.36
2026.04
34--
2026.05
33.51--
2026.05
32.98--
2026.05
29.84--
2026.05
28.8--
2026.05
28.27--
2026.05
27.75--
2026.05
27.23--
2026.05
27.23--
2026.05
25.65--
2026.05
21.99--