# Zero-Shot Relation Extraction via Reading Comprehension

## TL;DR

This paper presents a novel approach to relation extraction by framing it as a reading comprehension task, enabling zero-shot learning and leveraging large-scale neural models trained on crowd-sourced questions.

## Contribution

It introduces a method that reduces relation extraction to question answering, allowing zero-shot generalization and the use of large, diverse training datasets.

## Key findings

- High accuracy on known relation types
- Effective zero-shot relation extraction
- Large-scale training data improves performance

## Abstract

We show that relation extraction can be reduced to answering simple reading comprehension questions, by associating one or more natural-language questions with each relation slot. This reduction has several advantages: we can (1) learn relation-extraction models by extending recent neural reading-comprehension techniques, (2) build very large training sets for those models by combining relation-specific crowd-sourced questions with distant supervision, and even (3) do zero-shot learning by extracting new relation types that are only specified at test-time, for which we have no labeled training examples. Experiments on a Wikipedia slot-filling task demonstrate that the approach can generalize to new questions for known relation types with high accuracy, and that zero-shot generalization to unseen relation types is possible, at lower accuracy levels, setting the bar for future work on this task.