International Conference on Computational Linguistics and Intelligent Text Processing
One of the most basic functions of language is to refer to objects in a shared scene. Modeling reference with continuous representations is challenging because it requires individuation, i.e., tracking and distinguishing an arbitrary number of referents. We introduce a neural network model that, given a de nite description and a set of objects represented by natural images, points to the intended object if the expression has a unique referent, or indicates a failure, if it does not. The model, directly trained on reference acts, is competitive with a pipeline manually engineered to perform the same task, both when referents are purely visual, and when they are characterized by a combination of visual and linguistic properties.
Boleda G, Baroni M, Padó S. "Show me the cup": reference with continuous representations. In: Gelbukh, A. (ed.). International Conference on Computational Linguistics and Intelligent Text Processing. 1 ed. New York: Springer; 2018. p. 209-224.