Honest Students from Untrusted Teachers: Learning an Interpretable Question-Answering Pipeline from a Pretrained Language Model

arXiv:2210.02498v3 Announce Type: replace-cross
Abstract: Explainable question answering systems should produce not only accurate answers but also rationales that justify their reasoning and allow humans to check their work. But what sorts of rationales are useful and how can we train systems to produce them? We propose a new style of rationale for open-book question answering, called \emph{markup-and-mask}, which combines aspects of extractive and free-text explanations. In the markup phase, the passage is augmented with free-text markup that enables each …

abstract arxiv check cs.cl cs.lg humans language language model pipeline pretrained language model question question answering reasoning students systems teachers them train type work