Bootleg: Chasing the Tail with Self-Supervised Named Entity Disambiguation

TL;DR

Bootleg is a self-supervised named entity disambiguation system that leverages reasoning patterns inspired by human disambiguation strategies, achieving state-of-the-art results and transferring well to related tasks.

Contribution

The paper introduces Bootleg, a novel self-supervised NED approach grounded in explicit reasoning patterns, with a training procedure that enhances disambiguation of tail entities.

Findings

Bootleg outperforms existing NED benchmarks.

Learned representations transfer effectively to other entity-based tasks.

Achieves new state-of-the-art in TACRED relation extraction.

Abstract

A challenge for named entity disambiguation (NED), the task of mapping textual mentions to entities in a knowledge base, is how to disambiguate entities that appear rarely in the training data, termed tail entities. Humans use subtle reasoning patterns based on knowledge of entity facts, relations, and types to disambiguate unfamiliar entities. Inspired by these patterns, we introduce Bootleg, a self-supervised NED system that is explicitly grounded in reasoning patterns for disambiguation. We define core reasoning patterns for disambiguation, create a learning procedure to encourage the self-supervised model to learn the patterns, and show how to use weak supervision to enhance the signals in the training data. Encoding the reasoning patterns in a simple Transformer architecture, Bootleg meets or exceeds state-of-the-art on three NED benchmarks. We further show that the learned representations from Bootleg successfully transfer to other non-disambiguation tasks that require entity-based knowledge: we set a new state-of-the-art in the popular TACRED relation extraction task by 1.0 F1 points and demonstrate up to 8% performance lift in highly optimized production search and assistant tasks at a major technology company