Varifocal Question Generation for Fact-checking

TL;DR

This paper introduces Varifocal, a question generation method for fact-checking that uses different focal points within a claim and its metadata to produce more relevant questions, improving automatic and manual evaluation results.

Contribution

The paper presents Varifocal, a novel question generation approach leveraging focal points within claims and metadata, advancing fact-checking question generation.

Findings

Outperforms previous methods on fact-checking question datasets

Generates more relevant and informative questions according to manual evaluation

Demonstrates effectiveness in generating clarification questions for product descriptions

Abstract

Fact-checking requires retrieving evidence related to a claim under investigation. The task can be formulated as question generation based on a claim, followed by question answering. However, recent question generation approaches assume that the answer is known and typically contained in a passage given as input, whereas such passages are what is being sought when verifying a claim. In this paper, we present {\it Varifocal}, a method that generates questions based on different focal points within a given claim, i.e.\ different spans of the claim and its metadata, such as its source and date. Our method outperforms previous work on a fact-checking question generation dataset on a wide range of automatic evaluation metrics. These results are corroborated by our manual evaluation, which indicates that our method generates more relevant and informative questions. We further demonstrate the potential of focal points in generating sets of clarification questions for product descriptions.