A Gated Self-attention Memory Network for Answer Selection

TL;DR

This paper introduces a novel gated self-attention memory network for answer selection, outperforming previous models and achieving state-of-the-art results on TrecQA and WikiQA datasets.

Contribution

The paper proposes a new gated self-attention memory network architecture that departs from traditional compare-aggregate models for answer selection.

Findings

Outperforms previous methods significantly

Achieves state-of-the-art results on TrecQA and WikiQA datasets

Effective transfer learning from large-scale online corpus

Abstract

Answer selection is an important research problem, with applications in many areas. Previous deep learning based approaches for the task mainly adopt the Compare-Aggregate architecture that performs word-level comparison followed by aggregation. In this work, we take a departure from the popular Compare-Aggregate architecture, and instead, propose a new gated self-attention memory network for the task. Combined with a simple transfer learning technique from a large-scale online corpus, our model outperforms previous methods by a large margin, achieving new state-of-the-art results on two standard answer selection datasets: TrecQA and WikiQA.