The Impact of Visual Context on the Content of Referring Expressions
Henriette Anna Elisabeth Viethen, Robert Dale, Markus Guhe · Data Archiving and Networked Services (DANS) · 2011
Traditional approaches to referring expression generation (REG) have taken as a fundamental requirement the need to distinguish the intended referent from other entities in the context.It seems obvious that this should be a necessary condition for successful reference; but we suggest that a number of recent investigations cast doubt on the significance of this aspect of reference.In the present paper, we look at the role of visual context in determining the content of a referring expression, and come to the conclusion that, at least in the referential scenarios underlying our data, visual context appears not to be a major factor in content determination for reference.We discuss the implications of this surprising finding.