toplogo
Accedi
approfondimento - Identifying underlying capabilities and biases in vision-language model performance