Teaching Machines to Read and Comprehend

Karl Moritz Hermann; Tomáš Kočiský; Edward Grefenstette; Lasse Espeholt; Will Kay; Mustafa Suleyman; Phil Blunsom

doi:10.48550/arxiv.1506.03340

Abstract

Teaching machines to read natural language documents remains an elusive challenge. Machine reading systems can be tested on their ability to answer questions posed on the contents of documents that they have seen, but until now large scale training and test datasets have been missing for this type of evaluation. In this work we define a new methodology that resolves this bottleneck and provides large scale supervised reading comprehension data. This allows us to develop a class of attention based deep neural networks that learn to read real documents and answer complex questions with minimal prior knowledge of language structure.

Keywords

Computer scienceBottleneckReading (process)Reading comprehensionArtificial intelligenceClass (philosophy)Natural language processingScale (ratio)Test (biology)Question answeringComprehensionArtificial neural networkMachine learningLinguisticsProgramming language

Affiliated Institutions

Related Publications

Know What You Don’t Know: Unanswerable Questions for SQuAD

Pranav Rajpurkar , Robin Jia , Percy Liang

Extractive reading comprehension systems can often locate the correct answer to a question in a context document, but they also tend to make unreliable guesses on questions for ...

2018 Proceedings of the 56th Annual Meetin... 2087 citations

Natural Questions: A Benchmark for Question Answering Research

Tom Kwiatkowski , Jennimaria Palomaki , Olivia Redfield +15 more

We present the Natural Questions corpus, a question answering data set. Questions consist of real anonymized, aggregated queries issued to the Google search engine. An annotator...

2019 Transactions of the Association for C... 1830 citations

Analysis of polarity information in medical text.

Yun Niu , Xiaodan Zhu , Jianhua Li +1 more

Knowing the polarity of clinical outcomes is important in answering questions posed by clinicians in patient treatment. We treat analysis of this information as a classification...

2005 PubMed 83 citations

Enhancing Peer Interaction and Learning in the Classroom Through Reciprocal Questioning

Alison King

A guided reciprocal peer-questioning procedure was used by college students for learning expository material presented in classroom lectures. Students worked individually, using...

1990 American Educational Research Journal 352 citations

Large language models encode clinical knowledge

Karan Singhal , Shekoofeh Azizi , Tao Tu +29 more

Abstract Large language models (LLMs) have demonstrated impressive capabilities, but the bar for clinical applications is high. Attempts to assess the clinical knowledge of mode...

2023 Nature 2248 citations

Publication Info

Year: 2015
Type: preprint
Citations: 1527
Access: Closed

External Links

View on DOI.org

Social Impact

Altmetric

Teaching Machines to Read and Comprehend

PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

1527

OpenAlex

Cite This

APA Style

                            
                                    Karl Moritz Hermann, 
                                
                                    Tomáš Kočiský, 
                                
                                    Edward Grefenstette
                                
                                et al.
                            
                            (2015). 
                            Teaching Machines to Read and Comprehend. 
                            arXiv (Cornell University)
                            
                            .
                            https://doi.org/10.48550/arxiv.1506.03340

Identifiers

DOI: 10.48550/arxiv.1506.03340

Data Quality

Data completeness: 77%