How to model a pair of sentences is a critical issue in many NLP tasks such as answer selection (AS), paraphrase identification (PI) and textual entailment (TE). Most prior work (i) deals with one individual task by fine-tuning a specific system; (ii) models each sentence's representation separately, rarely considering the impact of the other sentence; or (iii) relies fully on manually designed, task-specific linguistic features. This work presents a general Attention Based Convolutional Neural Network (ABCNN) for modeling a pair of sentences. We make three contributions. (i) The ABCNN can be applied to a wide variety of tasks that require modeling of sentence pairs. (ii) We propose three attention schemes that integrate mutual influence between sentences into CNNs; thus, the representation of each sentence takes into consideration its counterpart. These interdependent sentence pair representations are more powerful than isolated sentence representations. (iii) ABCNNs achieve state-of-the-art performance on AS, PI and TE tasks. We release code at: https://github.com/ yinwenpeng/Answer_Selection.

Relation detection is a core component of many NLP applications including Knowledge Base Question Answering (KBQA). In this paper, we propose a hierarchical recurrent neural network enhanced by residual learning which detects KB relations given an input question. Our method uses deep residual bidirectional LSTMs to compare questions and relation names via different levels of abstraction. Additionally, we propose a simple KBQA system that integrates entity linking and our proposed relation detector to make the two components enhance each other. Our experimental results show that our approach not only achieves outstanding relation detection performance, but more importantly, it helps our KBQA system achieve state-of-the-art accuracy for both single-relation (SimpleQuestions) and multi-relation (WebQSP) QA benchmarks.

Zero-shot text classification (0SHOT-TC) is a challenging NLU problem to which little attention has been paid by the research community. 0SHOT-TC aims to associate an appropriate label with a piece of text, irrespective of the text domain and the aspect (e.g., topic, emotion, event, etc.) described by the label. And there are only a few articles studying 0SHOT-TC, all focusing only on topical categorization which, we argue, is just the tip of the iceberg in 0SHOT-TC. In addition, the chaotic experiments in literature make no uniform comparison, which blurs the progress.This work benchmarks the 0SHOT-TC problem by providing unified datasets, standardized evaluations, and state-of-the-art baselines. Our contributions include: i) The datasets we provide facilitate studying 0SHOT-TC relative to conceptually different and diverse aspects: the "topic" aspect includes "sports" and "politics" as labels; the "emotion" aspect includes "joy" and "anger"; the "situation" aspect includes "medical assistance" and "water shortage". ii) We extend the existing evaluation setup (labelpartially-unseen) -given a dataset, train on some labels, test on all labels -to include a more challenging yet realistic evaluation label-fully-unseen 0SHOT-TC (Chang et al., 2008), aiming at classifying text snippets without seeing task specific training data at all. iii) We unify the 0SHOT-TC of diverse aspects within a textual entailment formulation and study it this way. 1The plague in Mongolia, occurring last week, has caused more than a thousand isolation health,

We present a new deep learning architecture Bi-CNN-MI for paraphrase identification (PI). Based on the insight that PI requires comparing two sentences on multiple levels of granularity, we learn multigranular sentence representations using convolutional neural network (CNN) and model interaction features at each level. These features are then the input to a logistic classifier for PI. All parameters of the model (for embeddings, convolution and classification) are directly optimized for PI. To address the lack of training data, we pretrain the network in a novel way using a language modeling task. Results on the MSRP corpus surpass that of previous NN competitors.

Recently, Neural Network Language Models have been effectively applied to many types of Natural Language Processing (NLP) tasks. One popular type of tasks is the discovery of semantic and syntactic regularities that support the researchers in building a lexicon. Word embedding representations are notably good at discovering such linguistic regularities. We argue that two supervised learning approaches based on word embeddings can be successfully applied to the hypernym problem, namely, utilizing embedding offsets between word pairs and learning semantic projection to link the words. The offset-based model classifies offsets as hypernym or not. The semantic projection approach trains a semantic transformation matrix that ideally maps a hyponym to its hypernym. A semantic projection model can learn a projection matrix provided that there is a sufficient number of training word pairs. However, we argue that such models tend to learn is-a-particular-hypernym relation rather than to generalize is-a relation. The embeddings are trained by applying both the Continuous Bag-of Words and the Skip-Gram training models using a huge corpus in Turkish text. The main contribution of the study is the development of a novel and efficient architecture that is well-suited to applying word embeddings approaches to the Turkish language domain. We report that both the projection and the offset classification models give promising and novel results for the Turkish Language.