View article
Know What You Don’t Know: Unanswerable Questions for SQuAD Open
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 which the correct answer is not stated in the context. Existin…
View article
Natural Questions: A Benchmark for Question Answering Research Open
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 is presented with a question along with a Wikipedia page from…
View article
Bidirectional Attention Flow for Machine Comprehension Open
Machine comprehension (MC), answering a query about a given context paragraph, requires modeling complex interactions between the context and the query. Recently, attention mechanisms have been successfully extended to MC. Typically these …
View article
Identifying the PECO: A framework for formulating good questions to explore the association of environmental and other exposures with health outcomes Open
[First paragraph] A clearly-framed question creates the structure and delineates the approach to defining research objectives, conducting systematic reviews and developing health guidance (Guyatt et al., 2011; Armstrong et al., 2007). To a…
View article
Learning to Ask: Neural Question Generation for Reading Comprehension Open
We study automatic question generation for sentences from text passages in reading comprehension. We introduce an attention-based sequence learning model for the task and investigate the effect of encoding sentence- vs. paragraph-level inf…
View article
QAngaroo (MedHop + WikiHop) - Constructing Datasets for Multi-hop Reading Comprehension Across Documents Open
Most Reading Comprehension methods limit themselves to queries which can be answered using a single sentence, paragraph, or document. Enabling models to combine disjoint pieces of textual evidence would extend the scope of machine comprehe…
View article
Correction: Corrigendum: Salicylic acid biosynthesis is enhanced and contributes to increased biotrophic pathogen resistance in Arabidopsis hybrids Open
Nature Communications 6: Article number: 7309 (2015); Published online: 12 June 2015; Updated: 12 August 2015. In the last paragraph of the Discussion section in this Article, the hybrid Ler & C24 was incorrectly referred to as Ler & Col-0…
View article
Using ChatGPT to write patient clinic letters Open
The appropriate recording and communication of clinical information between clinicians and patients are of paramount importance. Recently there has been a much-needed drive to improve the information that is shared with patients.1General M…
View article
Bidirectional Attention Flow for Machine Comprehension Open
Machine comprehension (MC), answering a query about a given context paragraph, requires modeling complex interactions between the context and the query. Recently, attention mechanisms have been successfully extended to MC. Typically these …
View article
Correction: Corrigendum: Elucidating the role of disorder and free-carrier recombination kinetics in CH3NH3PbI3 perovskite films Open
Nature Communications 6: Article number: 8903 (2015); Published: 30 July 2015; Updated: 10 March 2016 This article contains errors in the units used for carrier mobility. In Fig. 2a–d, the units on the y axis should be ‘cm2V−1s−1’ not ‘V−1…
View article
Paragraph-level Neural Question Generation with Maxout Pointer and Gated Self-attention Networks Open
Question generation, the task of automatically creating questions that can be answered by a certain span of text within a given passage, is important for question-answering and conversational systems in digital assistants such as Alexa, Co…
View article
Document Embedding with Paragraph Vectors Open
Paragraph Vectors has been recently proposed as an unsupervised method for learning distributed representations for pieces of texts. In their work, the authors showed that the method can learn an embedding of movie review texts which can b…
View article
A Unified Model for Extractive and Abstractive Summarization using Inconsistency Loss Open
We propose a unified model combining the strength of extractive and abstractive summarization. On the one hand, a simple extractive model can obtain sentence-level attention with high ROUGE scores but less readable. On the other hand, a mo…
View article
Know What You Don't Know: Unanswerable Questions for SQuAD Open
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 which the correct answer is not stated in the context. Existin…
View article
A Hierarchical Neural Autoencoder for Paragraphs and Documents Open
Natural language generation of coherent long texts like paragraphs or longer documents is a challenging problem for recurrent networks models. In this paper, we explore an important step toward this generation task: training an LSTM (Long-…
View article
Advanced learners’ foreign language-related emotions across the four skills Open
Individual differences researchers have recently begun to investigate the concept of emotions and their role in language learning (MacIntyre, Gregersen, & Mercer, 2016). Our aim is to report on a project exploring English majors’ feelings …
View article
Dynamically Fused Graph Network for Multi-hop Reasoning Open
Text-based question answering (TBQA) has been studied extensively in recent years. Most existing approaches focus on finding the answer to a question within a single paragraph. However, many difficult questions require multiple supporting …
View article
Harvesting Paragraph-level Question-Answer Pairs from Wikipedia Open
We study the task of generating from Wikipedia articles question-answer pairs that cover content beyond a single sentence. We propose a neural network approach that incorporates coreference knowledge via a novel gating mechanism. As compar…
View article
MART: Memory-Augmented Recurrent Transformer for Coherent Video Paragraph Captioning Open
Generating multi-sentence descriptions for videos is one of the most challenging captioning tasks due to its high requirements for not only visual relevance but also discourse-based coherence across the sentences in the paragraph. Towards …
View article
Correction: Corrigendum: Ultrasensitive and label-free molecular-level detection enabled by light phase control in magnetoplasmonic nanoantennas Open
Nature Communications 6: Article number: 6150 (2015). Published 2 February 2015; Updated 18 June 2015. This Article contains typographical errors in the equations of the last sentence of the first paragraph of the Results section. This sen…
View article
Denoising Distantly Supervised Open-Domain Question Answering Open
Distantly supervised open-domain question answering (DS-QA) aims to find answers in collections of unlabeled text. Existing DS-QA models usually retrieve related paragraphs from a large-scale corpus and apply reading comprehension techniqu…
View article
Context-Aware Visual Policy Network for Fine-Grained Image Captioning Open
With the maturity of visual detection techniques, we are more ambitious in describing visual content with open-vocabulary, fine-grained and free-form language, i.e., the task of image captioning. In particular, we are interested in generat…
View article
Hierarchical Graph Network for Multi-hop Question Answering Open
In this paper, we present Hierarchical Graph Network (HGN) for multi-hop question answering. To aggregate clues from scattered texts across multiple paragraphs, a hierarchical graph is created by constructing nodes on different levels of g…
View article
Blockwise Self-Attention for Long Document Understanding Open
We present BlockBERT, a lightweight and efficient BERT model for better modeling long-distance dependencies. Our model extends BERT by introducing sparse block structures into the attention matrix to reduce both memory consumption and trai…
View article
Erratum to “The International Federation of Gynecology and Obstetrics (FIGO) initiative on pre‐eclampsia: A pragmatic guide for first‐trimester screening and prevention” [Int J Gynecol Obstet 145 Suppl. 1 (2019) 1–33] Open
The International Journal of Gynecology and Obstetrics regrets that, in the above article, an error appeared in the 5.3.4 Measurement of uterine artery pulsatility index section on page 17, 5th paragraph, second last sentence: The first-tr…
View article
Migrant Farmworkers Face Heightened Vulnerabilities During COVID-19 Open
First paragraph: The COVID-19 pandemic has dramatically reshaped Canadian society in just a few short weeks. At the same time, its varied impacts shine a light on pre-existing social inequities. Certain populations, including low wage work…
View article
Learning to Ask: Neural Question Generation for Reading Comprehension Open
We study automatic question generation for sentences from text passages in reading comprehension. We introduce an attention-based sequence learning model for the task and investigate the effect of encoding sentence- vs. paragraph-level inf…
View article
Tracking State Changes in Procedural Text: a Challenge Dataset and Models for Process Paragraph Comprehension Open
Bhavana Dalvi, Lifu Huang, Niket Tandon, Wen-tau Yih, Peter Clark. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers). 2018.
View article
A Two-Stage Parsing Method for Text-Level Discourse Analysis Open
Previous work introduced transition-based algorithms to form a unified architecture of parsing rhetorical structures (including span, nuclearity and relation), but did not achieve satisfactory performance. In this paper, we propose that tr…
View article
Pre-training Tasks for Embedding-based Large-scale Retrieval Open
We consider the large-scale query-document retrieval problem: given a query (e.g., a question), return the set of relevant documents (e.g., paragraphs containing the answer) from a large document corpus. This problem is often solved in two…