Multi-facet Universal Schema

TL;DR

This paper introduces a multi-facet universal schema model that represents sentence patterns with multiple embeddings to better capture their diverse facets, improving relation extraction accuracy and enabling entailment detection without manual labels.

Contribution

It proposes a neural multi-facet embedding approach for universal schema, addressing the limitations of single-facet models in relation extraction tasks.

Findings

Multi-facet embeddings outperform single-facet models in distantly supervised RE.

The approach enables entailment detection between sentence patterns without manual labels.

Significant improvement over compositional universal schema (CUSchema).

Abstract

Universal schema (USchema) assumes that two sentence patterns that share the same entity pairs are similar to each other. This assumption is widely adopted for solving various types of relation extraction (RE) tasks. Nevertheless, each sentence pattern could contain multiple facets, and not every facet is similar to all the facets of another sentence pattern co-occurring with the same entity pair. To address the violation of the USchema assumption, we propose multi-facet universal schema that uses a neural model to represent each sentence pattern as multiple facet embeddings and encourage one of these facet embeddings to be close to that of another sentence pattern if they co-occur with the same entity pair. In our experiments, we demonstrate that multi-facet embeddings significantly outperform their single-facet embedding counterpart, compositional universal schema (CUSchema) (Verga et al., 2016), in distantly supervised relation extraction tasks. Moreover, we can also use multiple embeddings to detect the entailment relation between two sentence patterns when no manual label is available.