Query-Reduction Networks for Question Answering
This addresses the challenge of multi-fact reasoning in question answering for AI systems, offering an incremental improvement with parallelization benefits.
The paper tackles the problem of question answering requiring reasoning over multiple facts by proposing Query-Reduction Network (QRN), a variant of RNN that handles short-term and long-term dependencies to reduce queries as it processes context, achieving state-of-the-art results on bAbI QA and dialog tasks and a real goal-oriented dialog dataset.
In this paper, we study the problem of question answering when reasoning over multiple facts is required. We propose Query-Reduction Network (QRN), a variant of Recurrent Neural Network (RNN) that effectively handles both short-term (local) and long-term (global) sequential dependencies to reason over multiple facts. QRN considers the context sentences as a sequence of state-changing triggers, and reduces the original query to a more informed query as it observes each trigger (context sentence) through time. Our experiments show that QRN produces the state-of-the-art results in bAbI QA and dialog tasks, and in a real goal-oriented dialog dataset. In addition, QRN formulation allows parallelization on RNN's time axis, saving an order of magnitude in time complexity for training and inference.