Many theories of semantic interpretation use λ-term manipulation to compositionally compute the meaning of a sentence. These theories are usually implemented in a language such as Prolog that can simulate λ-term operations with first-order unification. However, for some interesting cases, such as a Combinatory Categorial Grammar account of coordination constructs, this can only be done by obscu...

The idea of treating utterances as programs to be run in human brains or on a computer is jxirsued by advocating the use of the environment notion for natural language semantics. One structure of environments is devoted to keeping track of the real or fictive interlocutors and another to distinguishing somebody's linguistic behavior from its pretense and belief, while the third one represents t...

Extensible Markup Language documents are composed of sequences of character data and embedded markup, under control of a Document Type Definition. We will show that embedded markup needs to be dealt with at a syntactic level to account for interdependencies between linguistic structure and document structure as found in style guides and Controlled Languages. Furthermore, we discuss the effect o...

The 1980s have seen the advent of the first commercial applications of artificial intelligence research. The two prime examples are expert systems and natural language processors. These developments have already begun to affect the design of office computers. Future office workstations will have some 'understanding' of office procedures and of the goals of office workers and their organisation....

Comprehension and production of text and discourse do not solely depend on linguistic expressions, but also on the physical context. The questions addressed in this study are 1) whether deictic gestures are substitutable for deictic expressions, and 2) whether deictic gestures establish joint attention. An eye tracking experiment investigated the effect of referring expressions and gestures on ...

The task of any phonetic theory is to determine the form of a phonetic component by establishing the internal and external constraints on that component. The phonetic component itself converts linguistic knowledge of the structure of the speech act into time-varying commands suitable for control of the articulatory mechanism. Performing involves knowledge, and this knowledge must be expressed i...

The goal of this work is to investigate the acquisition of particular elementary linguistic functions in a simulation environment. The linguistic functions are denotation of objects and spatial relations. They are placed in the context of simulated action. A simulated micro-world, as it is known from research in situated action and learning (e.g. robotics), is supplemented by linguistic explana...

Many task domains require robots to interpret and act upon natural language commands which are given by people and which refer to the robot’s physical surroundings. Such interpretation is known variously as the symbol grounding problem (Harnad, 1990), grounded semantics (Feldman et al., 1996) and grounded language acquisition (Nenov and Dyer, 1993, 1994). This problem is challenging because peo...

We present a toolkit for manipulating and visualising time-aligned linguistic data such as dialogue transcripts or language processing data. The package complements existing editing tools by allowing for conversion between their formats, information extraction from the raw files, and by adding sophisticated, and easily extended methods for visualising the dynamics of dialogue processing. To ill...