Improving Entity Linking through Semantic Reinforced Entity Embeddings

TL;DR

This paper introduces FGS2EE, a method to enhance entity embeddings with semantic information, improving neural entity linking by making embeddings less distinctive and more contextually aligned, leading to state-of-the-art results.

Contribution

The paper proposes FGS2EE, a novel approach to inject semantic type information into entity embeddings, improving their contextual utility for entity linking.

Findings

Achieved state-of-the-art performance on entity linking benchmarks.

Demonstrated that semantic-enhanced embeddings improve contextual learning.

Validated effectiveness across extensive experiments.

Abstract

Entity embeddings, which represent different aspects of each entity with a single vector like word embeddings, are a key component of neural entity linking models. Existing entity embeddings are learned from canonical Wikipedia articles and local contexts surrounding target entities. Such entity embeddings are effective, but too distinctive for linking models to learn contextual commonality. We propose a simple yet effective method, FGS2EE, to inject fine-grained semantic information into entity embeddings to reduce the distinctiveness and facilitate the learning of contextual commonality. FGS2EE first uses the embeddings of semantic type words to generate semantic embeddings, and then combines them with existing entity embeddings through linear aggregation. Extensive experiments show the effectiveness of such embeddings. Based on our entity embeddings, we achieved new sate-of-the-art performance on entity linking.