Learning Knowledge-Enhanced Contextual Language Representations for Domain Natural Language Understanding

TL;DR

This paper introduces KANGAROO, a novel framework for learning knowledge-enhanced language representations tailored for closed-domain NLP tasks, effectively capturing implicit graph structures and addressing knowledge sparsity.

Contribution

The paper proposes a new method combining shallow relational triples and hyperbolic embeddings, along with contrastive learning on subgraphs, to improve KEPLMs in closed domains.

Findings

KANGAROO outperforms existing KEPLM methods in closed-domain NLP tasks.

The approach effectively handles knowledge sparsity and hierarchical structures.

Significant improvements in both full and few-shot learning settings.

Abstract

Knowledge-Enhanced Pre-trained Language Models (KEPLMs) improve the performance of various downstream NLP tasks by injecting knowledge facts from large-scale Knowledge Graphs (KGs). However, existing methods for pre-training KEPLMs with relational triples are difficult to be adapted to close domains due to the lack of sufficient domain graph semantics. In this paper, we propose a Knowledge-enhanced lANGuAge Representation learning framework for various clOsed dOmains (KANGAROO) via capturing the implicit graph structure among the entities. Specifically, since the entity coverage rates of closed-domain KGs can be relatively low and may exhibit the global sparsity phenomenon for knowledge injection, we consider not only the shallow relational representations of triples but also the hyperbolic embeddings of deep hierarchical entity-class structures for effective knowledge fusion.Moreover, as two closed-domain entities under the same entity-class often have locally dense neighbor subgraphs counted by max point biconnected component, we further propose a data augmentation strategy based on contrastive learning over subgraphs to construct hard negative samples of higher quality. It makes the underlying KELPMs better distinguish the semantics of these neighboring entities to further complement the global semantic sparsity. In the experiments, we evaluate KANGAROO over various knowledge-aware and general NLP tasks in both full and few-shot learning settings, outperforming various KEPLM training paradigms performance in closed-domains significantly.