Varifocal Question Generation for Fact-checking
This addresses the need for better evidence retrieval in fact-checking by generating questions without known answers, though it is incremental as it builds on existing question generation approaches.
The paper tackles the problem of generating questions for fact-checking by proposing Varifocal, a method that creates questions based on different focal points within a claim, such as spans and metadata, resulting in outperforming previous work on various automatic metrics and manual evaluations showing more relevant and informative questions.
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.