When humans and artificial agents (e.g. robots) have mismatched perceptions of the shared environment, referential communication between them becomes difficult. To mediate perceptual differences, this paper presents a new approach using probabilistic labeling for referential grounding. This approach aims to integrate different types of evidence from the collaborative referential discourse into ...

This thesis presents a model of incremental natural language understanding based on the grammatical formalism known as Combinatory Categorial Grammar. The model constitutes an integrated system involving a cyclical process of parsing, semantic adjudication and filtering. The motivating data for the model are the well-known observations about garden path effects in human sentence processing, and...

All forms of non-monotonic reasoning so far proposed suffer from a common failing: they are not computationally highly effective, indeed for the most part cannot be carried out even in principle. They inv olve either selfconsistency tests, or general theorem-proving techniques. Here we argue that, in fact, all non-monotonic logics proposed involve a kind of self-referential reasoning, and that ...

this study aims to understand certain linguistic and semantic resources for the text construction, namely the constructs of cohesion, coherence. the analysis of cohesive ties was conducted on the writing samples of 40 subjects (20 most coherent and 20 least coherent) iranian undergraduates of english. this prompted us to identify the dominant types of cohesive devices used in most coherent writ...

Most transformational linguists would no longer create explicit deep structures. Instead they adopt a surface-tnterpretlve approach. We find deep structures indIspensable for projection into a semantic network. In conjunction wrth a reference architecture based on constraint-posting, they mlnlmlze referential non-determinism% We extend Marcus’ Determm~sm Hypothesis to include rmmed/ate referenc...

Past studies show that a referential context is important for successful word learning. Still in question, however, is the success of word learning in the absence of a referential context. In this study 2-year-olds were presented with novel words in the absence of a speaker and therefore in the absence of a referential context. Findings revealed that word learning was successful across the expe...