EntEval: A Holistic Evaluation Benchmark for Entity Representations

TL;DR

EntEval introduces a comprehensive benchmark for evaluating entity representations across various tasks and proposes training methods leveraging Wikipedia hyperlinks to enhance these representations.

Contribution

The paper presents a new holistic evaluation benchmark for entity representations and develops training techniques using Wikipedia hyperlinks to improve entity modeling.

Findings

Improved performance on multiple EntEval tasks using new training objectives.

Demonstrated effectiveness of hyperlink-based training techniques.

Established a standardized benchmark for entity representation quality.

Abstract

Rich entity representations are useful for a wide class of problems involving entities. Despite their importance, there is no standardized benchmark that evaluates the overall quality of entity representations. In this work, we propose EntEval: a test suite of diverse tasks that require nontrivial understanding of entities including entity typing, entity similarity, entity relation prediction, and entity disambiguation. In addition, we develop training techniques for learning better entity representations by using natural hyperlink annotations in Wikipedia. We identify effective objectives for incorporating the contextual information in hyperlinks into state-of-the-art pretrained language models and show that they improve strong baselines on multiple EntEval tasks.