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

Visual Reasoning on VSP

83.7Accuracy

DeepLatent-RL-7B

9.319228.629647.9467.2504Jan 26, 2026Feb 15, 2026Mar 8, 2026Mar 29, 2026Apr 18, 2026May 9, 2026May 30, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
83.7
2026.01
78.36
2026.05
76
2026.01
71.36
2026.01
56.27
2026.01
55.64
2026.01
53.55
2026.01
45
2026.01
39.09
2026.01
35.09
2026.01
33.91
2026.01
30.45
2026.01
28.09
2026.01
26.73
2026.01
24.55
2026.05
13.5
2026.01
12.18