Resolving Indirect Referring Expressions for Entity Selection

TL;DR

This paper introduces a new dataset and models for resolving indirect referring expressions in natural language, improving entity selection in dialog systems with promising accuracy.

Contribution

It presents AltEntities, a large dataset of indirect references, and develops models that enhance understanding of natural indirect expressions for entity disambiguation.

Findings

Models achieve 82%-87% accuracy in disambiguation tasks.

New dataset enables research on indirect reference resolution.

Improves naturalness in dialog, recommendation, and search systems.

Abstract

Recent advances in language modeling have enabled new conversational systems. In particular, it is often desirable for people to make choices among specified options when using such systems. We address this problem of reference resolution, when people use natural expressions to choose between the entities. For example, given the choice `Should we make a Simnel cake or a Pandan cake?' a natural response from a dialog participant may be indirect: `let's make the green one'. Such natural expressions have been little studied for reference resolution. We argue that robustly understanding such language has large potential for improving naturalness in dialog, recommendation, and search systems. We create AltEntities (Alternative Entities), a new public dataset of 42K entity pairs and expressions (referring to one entity in the pair), and develop models for the disambiguation problem. Consisting of indirect referring expressions across three domains, our corpus enables for the first time the study of how language models can be adapted to this task. We find they achieve 82%-87% accuracy in realistic settings, which while reasonable also invites further advances.