Attention-over-Attention Neural Networks for Reading Comprehension
This addresses reading comprehension tasks for natural language processing, with incremental improvements in model architecture.
The paper tackles the problem of Cloze-style reading comprehension by proposing an attention-over-attention neural network model, which significantly outperforms state-of-the-art systems on public datasets like CNN and Children's Book Test.
Cloze-style queries are representative problems in reading comprehension. Over the past few months, we have seen much progress that utilizing neural network approach to solve Cloze-style questions. In this paper, we present a novel model called attention-over-attention reader for the Cloze-style reading comprehension task. Our model aims to place another attention mechanism over the document-level attention, and induces "attended attention" for final predictions. Unlike the previous works, our neural network model requires less pre-defined hyper-parameters and uses an elegant architecture for modeling. Experimental results show that the proposed attention-over-attention model significantly outperforms various state-of-the-art systems by a large margin in public datasets, such as CNN and Children's Book Test datasets.