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

Abstract Visual Reasoning on SVRT reformulated four-choice (test)

99.4Accuracy

PR-A2CL

21.71241.88162.0582.219Mar 1, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
99.486.7
2026.03
98.881
2026.03
98.579.3
2026.03
98.286.7
2026.03
96.986.7
2026.03
92.427.5
2026.03
92.157.9
2026.03
92.186.7
2026.03
89.355.7
2026.03
85.986.7
2026.03
83.569.2
2026.03
8353.8
2026.03
78.586.7
2026.03
70.330.8
2026.03
68.586.7
2026.03
68.238.2
2026.03
67.153.3
2026.03
65.281
2026.03
64.879.3
2026.03
5531.7
2026.03
5437.1
2026.03
51.338.3
2026.03
50.328.5
2026.03
4026.4
2026.03
2630.8
2026.03
25.828.5
2026.03
25.627.5
2026.03
24.726.4