Machine Comprehension Using Match-LSTM and Answer Pointer

By Shuohang Wang, Jing Jiang. 2016

600 citations recorded by the source corpus.

Machine comprehension of text is an important problem in natural language processing. A recently released dataset, the Stanford Question Answering Dataset (SQuAD), offers a large number of real questions and their answers created by humans through crowdsourcing. SQuAD provides a challenging testbed for evaluating machine comprehension algorithms, partly because compared with previous datasets, in SQuAD the answers do not come from a small set of candidate answers and they have variable lengths. We propose an end-to-end neural architecture for the task. The architecture is based on match-LSTM, a model we proposed previously for textual entailment, and Pointer Net, a sequence-to-sequence model proposed by Vinyals et al.(2015) to constrain the output tokens to be from the input sequences. We propose two ways of using Pointer Net for our task. Our experiments show that both of our two models substantially outperform the best results obtained by Rajpurkar et al.(2016) using logistic regression and manually crafted features.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

Multi-Granularity Hierarchical Attention Fusion Networks for Reading Comprehension and Question Answering (2018, methodology): “AddSent AddOneSent Logistic (Rajpurkar et al., 2016) 23.2 30.4 Match-S (Wang and Jiang, 2016) 27.3 39.0 Match-E (Wang and Jiang, 2016) 29.4 41.8 BiDAF-S (Seo et al., 2016) 34.3 45.7 BiDAF-E (Seo et al., 2016) 34.2 46.9 ReasoNet-S (Shen et al., 2017) 39.4 50.3 ReasoNet-E (Shen et al., 2017) 39.4 49.8”
U-Net: Machine Reading Comprehension with Unanswerable Questions (2018, methodology): “Then we feed c q into the answer pointer to ﬁnd boundaries of answers (Wang and Jiang 2016), and the classiﬁcation layer to distinguish whether the question is answerable.”
Simple and Effective Multi-Paragraph Reading Comprehension (2017, background): “Neural reading comprehension systems typically use some form of attention (Wang and Jiang, 2016), although alternative architectures exist (Chen et al., 2017; Weissenborn et al., 2017b).”
The Natural Language Decathlon: Multitask Learning as Question Answering (2018, methodology): “This allowed deterministic answer extraction to overtake sequential token generation (Wang and Jiang, 2017).”
NewsQA: A Machine Comprehension Dataset (2016, methodology): “The first neural model is the match-LSTM (mLSTM) system of Wang & Jiang (2016b).”
Evidence Aggregation for Answer Re-Ranking in Open-Domain Question Answering (2017, methodology): “These studies adopt a two-step process: an information retrieval (IR) model to coarsely select passages relevant to a question, followed by a reading comprehension (RC) model (Wang & Jiang, 2017; Seo et al., 2017; Chen et al., 2017) to infer an answer from the passages.”
Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answering (2019, background): “Fourth , most effective QA and RC models highly rely on explicit inter-sentence matching between questions and passages (Wang and Jiang, 2017; Wang et al., 2016; Seo et al., 2017; Wang et al., 2017), whereas BERT only applies self-attention layers over the concatenation of a question-passage pair.”
Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering (2019, methodology): “For example, in R 3 (Wang et al., 2018a), the retriever uses Match-LSTM model (Wang & Jiang, 2017) to compute question-matching representations for each token in the passage.”
The NarrativeQA Reading Comprehension Challenge (2017, methodology): “Models which do well on SQuAD (e.g. Wang and Jiang (2016), Weissenborn et al. (2017)), extracting spans or pointing, do well here too, and the same concerns as above about the general applicability of solutions to this dataset to larger reading comprehension problems applies.”
Learning Recurrent Span Representations for Extractive Question Answering (2016, methodology): “In our experiments, we show an increase in performance over Wang & Jiang (2016) of 5% in terms of exact match to a reference answer, and 3 .”
Semi-Supervised QA with Generative Domain-Adaptive Nets (2017, background): “Various neural models based on attention mechanisms (Wang and Jiang, 2016; Seo et al., 2016; Xiong et al., 2016; Wang et al., 2016; Dhingra et al., 2016; Kadlec et al., 2016; Trischler et al., 2016b; Sordoni et al., 2016; Cui et al., 2016; Chen et al., 2016) have been proposed to tackle the tasks of”
Adversarial Examples for Evaluating Reading Comprehension Systems (2017, methodology): “Match-LSTM has a similar structure, but also has a hard-coded rule that stops it from predicting very long answers.”
Gated Self-Matching Networks for Reading Comprehension and Question Answering (2017, methodology): “Wang and Jiang (2016b) build question-aware passage representation with match-LSTM (Wang and Jiang, 2016a), and predict answer boundaries in the passage with pointer networks (Vinyals et al., 2015).”
Dynamic Coattention Networks For Question Answering (2016, methodology): “For SQuAD, Wang & Jiang (2016b) proposed an end-to-end neural network model that consists of a Match-LSTM encoder, originally introduced in Wang & Jiang (2016a), and a pointer network decoder (Vinyals et al., 2015); Yu et al. (2016) introduced a dynamic chunk reader, a neural reading comprehension…”
ReasoNet: Learning to Stop Reading in Machine Comprehension (2016, neutral): “…[18] 40.4 51.0 - - Dynamic Chunk Reader [34] 62.5 71.0 - - Fine-Grained Gating [33] 62.5 73.3 - - Match-LSTM [26] 64.7 73.7 67.9 77.0 RaSoR [14] - - 67.4 75.5 Multi-Perspective Matching [28] 68 Iterative Co-attention Network α 67.5 76.8 - - FastQA [30] 68.4…”
Making Neural QA as Simple as Possible but not Simpler (2017, methodology): “Since this was only an initial experiment on generative QA using extractive QA and the methodology used for training, pre-and post-processing on this dataset for the other models, especially for Wang and Jiang (2017), is unclear, the comparability to the other QA systems is limited.”
Multi-Perspective Context Matching for Machine Comprehension (2016, methodology): “Wang and Jiang (2016) proposed a match-LSTM model to match the passage with the question, then the Pointer Network (Vinyals et al., 2015) was utilized to select a list of positions from the passage as the ﬁnal answer (“Match-LSTM (Sequence)” in Table 2).”