In this section we discuss some important extensions of the uniform approach introduced in this thesis. We start by describing how Explanation-based Learning can be integrated with the uniform tabular algorithm to speed up processing. In [Neumann1994] we already reported on an implementation of such approach, however, only for the case of parsing. In the next section, we outline how this method can also be applied during generation.
In the section that follows we discuss the integration of preferences, processing of elliptical utterances, and fully incremental text processing. Finally, we discuss the relevance of the new uniform computational model under the perspective of cognitive processing.