Learning to Search in Long Documents Using Document Structure
arXiv:1806.03529
Abstract
Reading comprehension models are based on recurrent neural networks that sequentially process the document tokens. As interest turns to answering more complex questions over longer documents, sequential reading of large portions of text becomes a substantial bottleneck. Inspired by how humans use document structure, we propose a novel framework for reading comprehension. We represent documents as trees, and model an agent that learns to interleave quick navigation through the document tree with more expensive answer extraction. To encourage exploration of the document tree, we propose a new algorithm, based on Deep Q-Network (DQN), which strategically samples tree nodes at training time. Empirically we find our algorithm improves question answering performance compared to DQN and a strong information-retrieval (IR) baseline, and that ensembling our model with the IR baseline results in further gains in performance.
COLING 2018 (camera ready version); v2: added acknowledgments
References in corpus (7)
- Multiple Object Recognition with Visual Attention
- Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learning
- Quasi-Recurrent Neural Networks
- Evidence Aggregation for Answer Re-Ranking in Open-Domain Question Answering
- Learning Recurrent Span Representations for Extractive Question Answering
- Neural Speed Reading via Skim-RNN
- Question Answering from Unstructured Text by Retrieval and Comprehension