Learning to Select the Next Reasonable Mention for Entity Linking

TL;DR

This paper introduces DyMen, a reinforcement learning-based model that dynamically selects the next mention for entity linking, improving the utilization of previously linked entities and enhancing overall linking accuracy.

Contribution

The paper proposes a novel reinforcement learning approach for dynamic mention selection in entity linking, addressing limitations of fixed mention order in previous methods.

Findings

DyMen outperforms existing models on benchmark datasets.

Dynamic mention selection improves linking accuracy.

Sampling with sliding window maintains semantic coherence.

Abstract

Entity linking aims to establish a link between entity mentions in a document and the corresponding entities in knowledge graphs (KGs). Previous work has shown the effectiveness of global coherence for entity linking. However, most of the existing global linking methods based on sequential decisions focus on how to utilize previously linked entities to enhance the later decisions. In those methods, the order of mention is fixed, making the model unable to adjust the subsequent linking targets according to the previously linked results, which will cause the previous information to be unreasonably utilized. To address the problem, we propose a novel model, called DyMen, to dynamically adjust the subsequent linking target based on the previously linked entities via reinforcement learning, enabling the model to select a link target that can fully use previously linked information. We sample mention by sliding window to reduce the action sampling space of reinforcement learning and maintain the semantic coherence of mention. Experiments conducted on several benchmark datasets have shown the effectiveness of the proposed model.