Neural heuristics for scaling constructional language processing

Abstract

Constructionist approaches to language make use of form-meaning pairings, called constructions, to capture all linguistic knowledge that is necessary for comprehending and producing natural language expressions. Language processing consists then in combining the constructions of a grammar in such a way that they solve a given language comprehension or production problem. Finding such an adequate sequence of constructions constitutes a search problem that is combinatorial in nature and becomes intractable as grammars increase in size. In this paper, we introduce a neural methodology for learning heuristics that substantially optimise the search processes involved in constructional language processing. We validate the methodology in a case study for the CLEVR benchmark dataset. We show that our novel methodology outperforms state-of-the-art techniques in terms of size of the search space and time of computation, most markedly in the production direction. The results reported on in this paper have the potential to overcome the major efficiency obstacle that hinders current efforts in learning large-scale construction grammars, thereby contributing to the development of scalable constructional language processing systems.

Modelling constructivist language acquisition through syntactico-semantic pattern finding

“…If multiple combinations of constructions can apply to the same input utterance or meaning representation, the combination with the highest average entrenchment score will be preferred. We refer the interested reader to [ 89 ] for a detailed discussion of how constructional language processing is concretely operationalized in FCG.…”