Ask what's missing and what's useful: Improving Clarification Question Generation using Global Knowledge

The ability to generate clarification questions i.e., questions that identify\nuseful missing information in a given context, is important in reducing\nambiguity. Humans use previous experience with similar contexts to form a\nglobal view and compare it to the given context to ascertain what is missing\nand what is useful in the context. Inspired by this, we propose a model for\nclarification question generation where we first identify what is missing by\ntaking a difference between the global and the local view and then train a\nmodel to identify what is useful and generate a question about it. Our model\noutperforms several baselines as judged by both automatic metrics and humans.\n