Machine Reading as Model Construction

1 WHAT IS MACHINE READING? With the advent of large datasets of paragraphs + questions, e.g., SQuAD [4], TriviaQA [3], there has been renewed interest in generalpurpose “reading comprehension” (RC) systems, capable of answering questions against those paragraphs, e.g., [5, 6]. These systems have become remarkably effective at factoid QA. However, they require extensive training data, and can still struggle with queries requiring complex inference [1]. The extent to which these systems have truely read and understood the paragraph remains unclear [2]. At the other end of the spectrum, AI has also developed sophisticated formalisms for modeling the world, e.g., situation calculus, event calculus, qualitative modeling. These frameworks allow systems to represent facts which are known, and infer facts which are unknown. Models built with these frameworks constitute an understanding of the world, in that sense that they are predictive: If the model’s computational clockwork moves in a way similar to the world, then the model can predict how the world will behave, constituting a degree of understanding of the world. In this context, machine reading can be viewed as the task of constructing such models from text, given a particular modeling framework in which to express those models. While it is possible that a neural system might eventually be able to infer a predictive, neural model of the world solely from large numbers of examples, we do not believe this is likely in the near future. Rather, we see the way forward as combining the pattern-learning techniques of neural systems with the modeling capabilities of structured representations. AI modeling frameworks provide a set of primitives for constructing predictive models, and neural systems can help construct models within those frameworks that best fit data. The grand challenge for machine reading, going forward, is combining these two technologies together to do this.