Exploring Neural Entity Representations for Semantic Information
This work provides a unified evaluation framework for neural entity representations, which is important for researchers and practitioners in natural language processing to understand the strengths and weaknesses of different embedding methods.
This paper evaluates eight neural entity embedding methods using a suite of probing tasks and two entity linking tasks. It demonstrates how well these methods remember descriptive words, learn type, relationship, and factual information, and identify mention frequency, while also comparing their generalization across different architectures and datasets.
Neural methods for embedding entities are typically extrinsically evaluated on downstream tasks and, more recently, intrinsically using probing tasks. Downstream task-based comparisons are often difficult to interpret due to differences in task structure, while probing task evaluations often look at only a few attributes and models. We address both of these issues by evaluating a diverse set of eight neural entity embedding methods on a set of simple probing tasks, demonstrating which methods are able to remember words used to describe entities, learn type, relationship and factual information, and identify how frequently an entity is mentioned. We also compare these methods in a unified framework on two entity linking tasks and discuss how they generalize to different model architectures and datasets.