Neural Text Understanding with Attention Sum Reader

نویسندگان

  • Rudolf Kadlec
  • Martin Schmid
  • Ondřej Bajgar
  • Jan Kleindienst
چکیده

Two large-scale cloze-style context-question-answer datasets have been introduced recently: i) the CNN and Daily Mail news data and ii) the Children’s Book Test. Thanks to the size of these datasets, the associated task is well suited for deep-learning techniques that seem to outperform all alternative approaches. We present a new, simple model that is tailor made for such question-answering problems. Our model directly sums attention over candidate answer words in the document instead of using it to compute weighted sum of word embeddings. Our model outperforms models previously proposed for these tasks by a large margin.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Emergent Predication Structure in Vector Representations of Neural Readers

Reading comprehension is a question answering task where the answer is to be found in a given passage about entities and events not mentioned in general knowledge sources. A significant number of neural architectures for this task (neural readers) have recently been developed and evaluated on large cloze-style datasets. We present experiments supporting the emergence of “predication structure” ...

متن کامل

Char2wav: End-to-end Speech Synthesis

We present Char2Wav, an end-to-end model for speech synthesis. Char2Wav has two components: a reader and a neural vocoder. The reader is an encoderdecoder model with attention. The encoder is a bidirectional recurrent neural network that accepts text or phonemes as inputs, while the decoder is a recurrent neural network (RNN) with attention that produces vocoder acoustic features. Neural vocode...

متن کامل

Text Understanding with the Attention Sum Reader Network

Several large cloze-style context-questionanswer datasets have been introduced recently: the CNN and Daily Mail news data and the Children’s Book Test. Thanks to the size of these datasets, the associated text comprehension task is well suited for deep-learning techniques that currently seem to outperform all alternative approaches. We present a new, simple model that uses attention to directly...

متن کامل

Metadiscourse Markers Revisited in EFL Context: The Case of Iranian Academic Learners’ Perception of Written Texts

Moving in line with the postulation that metadiscourse (MD) markers help transform a dry and tortuous piece of text into a coherent and reader-friendly one, the researchers in the current study attempted to investigate the effect different metadiscourse markers might have on Iranian EFL learners’ perception of written texts. To this end, 120 undergraduate English students were given three diffe...

متن کامل

An LSTM Attention-based Network for Reading Comprehension

A significant goal of natural language processing (NLP) is to devise a system capable of machine understanding of text. A typical system can be tested on its ability to answer questions based on a given context document. One appropriate dataset for such a system is the Stanford Question Answering Dataset (SQuAD), a crowdsourced dataset of over 100k (question, context, answer) triplets. In this ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2016