Hierarchical Question Answering for Long Documents
arXiv:1611.01839
Abstract
We present a framework for question answering that can efficiently scale to longer documents while maintaining or even improving performance of state-of-the-art models. While most successful approaches for reading comprehension rely on recurrent neural networks (RNNs), running them over long documents is prohibitively slow because it is difficult to parallelize over sequences. Inspired by how people first skim the document, identify relevant parts, and carefully read these parts to produce an answer, we combine a coarse, fast model for selecting relevant sentences and a more expensive RNN for producing the answer from those sentences. We treat sentence selection as a latent variable trained jointly from the answer only using reinforcement learning. Experiments demonstrate the state of the art performance on a challenging subset of the Wikireading and on a new dataset, while speeding up the model by 3.5x-6.7x.
References in corpus (18)
- Hierarchical Question-Image Co-Attention for Visual Question Answering
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- Multiple Object Recognition with Visual Attention
- The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations
- Dynamic Coattention Networks For Question Answering
- A Convolutional Neural Network for Modelling Sentences
- Deep Learning for Answer Sentence Selection
- Attentive Pooling Networks
- Neural Summarization by Extracting Sentences and Words
- A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task
- Learning to Compose Neural Networks for Question Answering
- Multi-Perspective Context Matching for Machine Comprehension
- Language Understanding for Text-based Games Using Deep Reinforcement Learning
- Query-Reduction Networks for Question Answering
- NewsQA: A Machine Comprehension Dataset
- Improving Information Extraction by Acquiring External Evidence with Reinforcement Learning
- Who did What: A Large-Scale Person-Centered Cloze Dataset
- A Parallel-Hierarchical Model for Machine Comprehension on Sparse Data