This paper describes two supervised baseline systems for the Story Cloze Test Shared Task (Mostafazadeh et al., 2016a). We first build a classifier using features based on word embeddings and semantic similarity computation. We further implement a neural LSTM system with different encoding strategies that try to model the relation between the story and the provided endings. Our experiments show...

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...

In this paper we study the problem of answering cloze-style questions over documents. Our model, the Gated-Attention (GA) Reader1, 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...

Efficient language processing may involve generating expectations about upcoming input. To investigate the extent to which prediction might facilitate reading, a large-scale survey provided cloze scores for all 2689 words in 55 different text passages. Highly predictable words were quite rare (5% of content words), and most words had a more-expected competitor. An eye-tracking study showed sens...

When younger and older adults were allowed to adjust the speech rate of time-compressed and time-expanded speech passages, older adults tended to select as preferred rates significantly slower speech rates than the younger adults. Both age groups, however, selected slower rates for difficult speech passages (low cloze predictability) than for easy passages (high cloze predictability). Recall pe...

Traditional symbolic reasoning engines, while attractive for their precision and explicability, have a few major drawbacks: the use of brittle inference procedures that rely on exact matching (unification) logical terms, an inability to deal with uncertainty, need precompiled rule-base knowledge (the “knowledge acquisition” problem). To address these issues, we devise novel reasoner called Brai...

According to embodied theories, the symbols used by language are meaningful because they are grounded in perception, action, and emotion. In contrast, according to abstract symbol theories, meaning arises from the syntactic combination of abstract, amodal symbols. If language is grounded in internal bodily states, then one would predict that emotion affects language. Consistent with this, advoc...