Gated-Attention Readers for Text Comprehension

In this paper we study the problem of answering cloze-style questions over documents. Our model, the Gated-Attention (GA) Reader, integrates a multi-hop architecture with a novel attention mechanism, which is based on multiplicative interactions between the query embedding and the intermediate states of a recurrent neural network document reader. This enables the reader to build query-specific representations of tokens in the document for accurate answer selection. The GA Reader obtains state-of-the-art results on three benchmarks for this task--the CNN \& Daily Mail news stories and the Who Did What dataset. The effectiveness of multiplicative interaction is demonstrated by an ablation study, and by comparing to alternative compositional operators for implementing the gated-attention. The code is available at https://github.com/bdhingra/ga-reader.

PDF Abstract ACL 2017 PDF ACL 2017 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Question Answering Children's Book Test NSE Accuracy-CN 71.9% # 1
Accuracy-NE 73.2% # 2
Question Answering Children's Book Test GA reader Accuracy-CN 69.4% # 3
Accuracy-NE 71.9% # 5
Question Answering Children's Book Test GA + feature + fix L(w) Accuracy-CN 70.7% # 2
Accuracy-NE 74.9% # 1
Question Answering CNN / Daily Mail GA Reader CNN 77.9 # 2
Daily Mail 80.9 # 1
Open-Domain Question Answering Quasar GA EM (Quasar-T) 26.4 # 5
F1 (Quasar-T) 26.4 # 6

Methods


No methods listed for this paper. Add relevant methods here