Comparatives, Quantifiers, Proportions: a Multi-Task Model for the Learning of Quantities from Vision | Publicación