Question Answering is a Format; When is it Useful?

TL;DR

The paper argues that question answering should be viewed as a versatile format rather than a specific task, emphasizing its appropriate use cases and scope in research and dataset design.

Contribution

It clarifies the conceptual scope of question answering as a format and discusses when it is appropriately used in various tasks and datasets.

Findings

Question answering is best seen as a format, not a standalone task.

Proper framing of question answering enhances clarity in research.

The utility of question answering varies across different phenomena.

Abstract

Recent years have seen a dramatic expansion of tasks and datasets posed as question answering, from reading comprehension, semantic role labeling, and even machine translation, to image and video understanding. With this expansion, there are many differing views on the utility and definition of "question answering" itself. Some argue that its scope should be narrow, or broad, or that it is overused in datasets today. In this opinion piece, we argue that question answering should be considered a format which is sometimes useful for studying particular phenomena, not a phenomenon or task in itself. We discuss when a task is correctly described as question answering, and when a task is usefully posed as question answering, instead of using some other format.