A Gated Self-attention Memory Network for Answer Selection
This work addresses answer selection for applications like question-answering systems, representing an incremental advance by introducing a novel method for a known bottleneck.
The paper tackles answer selection by proposing a gated self-attention memory network that departs from the Compare-Aggregate architecture, achieving new state-of-the-art results on TrecQA and WikiQA datasets with a large margin improvement.
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.