Ellipsis and Coreference Resolution as Question Answering

Coreference (‘he eats potatoes’) and many forms of ellipsis (e.g., ‘so does Mary’) are similar to reading comprehension questions (‘who eats potatoes’ or ‘what does Mary do with potatoes’), in that in order to resolve these, we need to identify an appropriate text span in the previous discourse. This paper exploits this analogy and proposes to use an architecture developed for machine comprehension for ellipsis and coreference resolution. We present both single-task and joint models and evaluate them across standard benchmarks, outperforming the current state of the art for ellipsis by up to 48.5% error reduction – and for coreference by 37.5% error reduction.