Enhancing Language Representation with Constructional Information for Natural Language Understanding

TL;DR

This paper introduces a novel framework that integrates construction grammar with pre-trained language models to improve natural language understanding by capturing form-meaning pairings and high-order interactions.

Contribution

It proposes the HyCxG framework combining construction extraction, discriminative selection, and hypergraph attention to enrich language representations in NLU tasks.

Findings

Outperforms baseline models on multiple NLU tasks

Effectively captures high-order word interactions

Enhances the understanding of constructions in language models

Abstract

Natural language understanding (NLU) is an essential branch of natural language processing, which relies on representations generated by pre-trained language models (PLMs). However, PLMs primarily focus on acquiring lexico-semantic information, while they may be unable to adequately handle the meaning of constructions. To address this issue, we introduce construction grammar (CxG), which highlights the pairings of form and meaning, to enrich language representation. We adopt usage-based construction grammar as the basis of our work, which is highly compatible with statistical models such as PLMs. Then a HyCxG framework is proposed to enhance language representation through a three-stage solution. First, all constructions are extracted from sentences via a slot-constraints approach. As constructions can overlap with each other, bringing redundancy and imbalance, we formulate the conditional max coverage problem for selecting the discriminative constructions. Finally, we propose a relational hypergraph attention network to acquire representation from constructional information by capturing high-order word interactions among constructions. Extensive experiments demonstrate the superiority of the proposed model on a variety of NLU tasks.