Towards an Indexical Model of Situated Language Comprehension for Cognitive Agents in Physical Worlds

We propose a computational model of situated language comprehension based on\nthe Indexical Hypothesis that generates meaning representations by translating\namodal linguistic symbols to modal representations of beliefs, knowledge, and\nexperience external to the linguistic system. This Indexical Model incorporates\nmultiple information sources, including perceptions, domain knowledge, and\nshort-term and long-term experiences during comprehension. We show that\nexploiting diverse information sources can alleviate ambiguities that arise\nfrom contextual use of underspecific referring expressions and unexpressed\nargument alternations of verbs. The model is being used to support linguistic\ninteractions in Rosie, an agent implemented in Soar that learns from\ninstruction.\n