Unsupervised Relation Extraction from Language Models using Constrained Cloze Completion

TL;DR

This paper introduces RE-Flex, a simple unsupervised relation extraction framework using constrained cloze completion on pretrained language models, achieving significant improvements over existing methods without fine-tuning.

Contribution

The paper presents RE-Flex, a novel framework leveraging constrained inference on language models for unsupervised relation extraction, outperforming previous approaches.

Findings

RE-Flex outperforms competing methods by up to 27.8 F1 points.

Constrained inference enables accurate relation extraction without fine-tuning.

Extensive experiments validate the effectiveness of RE-Flex across multiple benchmarks.

Abstract

We show that state-of-the-art self-supervised language models can be readily used to extract relations from a corpus without the need to train a fine-tuned extractive head. We introduce RE-Flex, a simple framework that performs constrained cloze completion over pretrained language models to perform unsupervised relation extraction. RE-Flex uses contextual matching to ensure that language model predictions matches supporting evidence from the input corpus that is relevant to a target relation. We perform an extensive experimental study over multiple relation extraction benchmarks and demonstrate that RE-Flex outperforms competing unsupervised relation extraction methods based on pretrained language models by up to 27.8 $F_1$ points compared to the next-best method. Our results show that constrained inference queries against a language model can enable accurate unsupervised relation extraction.