Learning to Organize Knowledge and Answer Questions with N-Gram Machines

TL;DR

This paper introduces N-Gram Machines, a scalable approach for knowledge organization and question answering that uses symbolic representations to handle large texts efficiently, demonstrated on synthetic and real datasets.

Contribution

The paper presents N-Gram Machines, a novel method that improves scalability of question answering systems using symbolic indexing, enabling handling of large corpora with weak supervision.

Findings

Successfully solves synthetic bAbI tasks

Scales to millions of sentences in 'life-long bAbI'

Answers questions from Wikipedia with weak supervision

Abstract

Though deep neural networks have great success in natural language processing, they are limited at more knowledge intensive AI tasks, such as open-domain Question Answering (QA). Existing end-to-end deep QA models need to process the entire text after observing the question, and therefore their complexity in responding a question is linear in the text size. This is prohibitive for practical tasks such as QA from Wikipedia, a novel, or the Web. We propose to solve this scalability issue by using symbolic meaning representations, which can be indexed and retrieved efficiently with complexity that is independent of the text size. We apply our approach, called the N-Gram Machine (NGM), to three representative tasks. First as proof-of-concept, we demonstrate that NGM successfully solves the bAbI tasks of synthetic text. Second, we show that NGM scales to large corpus by experimenting on "life-long bAbI", a special version of bAbI that contains millions of sentences. Lastly on the WikiMovies dataset, we use NGM to induce latent structure (i.e. schema) and answer questions from natural language Wikipedia text, with only QA pairs as weak supervision.