HyperAI

Visual Commonsense Tests On Vicomte Color

Metrics

Spearman's Rho

Results

Performance results of various models on this benchmark

Comparison Table
Model NameSpearman's Rho
z-lavi-zero-shot-language-solver-fueled-by49.6
visual-commonsense-in-pretrained-unimodal-and-137.6