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

Visual Reasoning on MMStar 1.0 (test)

63.63Coarse Score

LLaVA-1.5-7B

26.50236.14145.7855.419Aug 25, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.08
63.6325.6338.8928.9226.618.4833.69
2025.08
59.5625.7140.4929.9427.9217.3733.5
2025.08
59.0822.6639.5129.6628.3920.6633.33
2025.08
56.3519.536.7227.7326.718.9430.99
2025.08
56.3221.5839.4830.2223.9815.1631.12
2025.08
55.2722.9239.3227.9524.7624.6232.47
2025.08
54.8221.0737.0327.8224.1819.3230.71
2025.08
54.1121.2635.0929.5620.3415.0429.23
2025.08
49.9921.4538.3728.4521.5418.5829.73
2025.08
48.5819.0439.7329.6922.9121.9530.32
2025.08
47.2521.0533.8925.7520.3216.7627.5
2025.08
43.7621.3432.5825.9723.1819.9627.8
2025.08
31.8519.123.7723.6114.2816.1121.45
2025.08
27.9321.2625.7121.8420.1817.8322.46