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

Vision-Language Benchmarking on MMBench (test)

80.19Logical Score

SSL4RL-7B Jigsaw

76.34277.34178.3479.339Oct 18, 2025
Updated 15d ago

Evaluation Results

MethodLinks
2025.10
80.1984.6488.0285.784.5390.9387.73
2025.10
79.0682.1385.2985.0285.4288.9586.25
2025.10
78.786.1687.1385.9188.4788.9287.5
2025.10
77.7383.1984.5885.8985.2987.7486.25
2025.10
76.4984.6885.6984.6684.4989.1586.37