Hidden variable models are important tools for solving open domain machine comprehension tasks and have achieved remarkable accuracy in many question answering benchmark datasets. Existing models impose strong independence assumptions on hidden variables, which leaves the interaction among them unexplored. Here we introduce linguistic structures to help capturing global evidence in hidden variable modeling. In the proposed algorithms, question-answer pairs are scored based on structured inference results on parse trees and semantic frames, which aims to assign hidden variables in a global optimal way. Experiments on the MCTest dataset demonstrate that the proposed models are highly competitive with state-of-the-art machine comprehension systems.

The machine reading task, where a computer reads a document and answers questions about it, is important in artificial intelligence research. Recently, many models have been proposed to address it. Word-level models, which have words as units of input and output, have proven to yield state-of-the-art results when evaluated on English datasets. However, in morphologically richer languages, many more unique words exist than in English due to highly productive prefix and suffix mechanisms. This may set back word-level models, since vocabulary sizes too big to allow for efficient computing may have to be employed. Multiple alternative input granularities have been proposed to avoid large input vocabularies, such as morphemes, character n-grams, and bytes. Bytes are advantageous as they provide a universal encoding format across languages, and allow for a small vocabulary size, which, moreover, is identical for every input language. In this work, we investigate whether bytes are suitable as input units across morphologically varied languages. To test this, we introduce two large-scale machine reading datasets in morphologically rich languages, Turkish and Russian. We implement 4 byte-level models, representing the major types of machine reading models and introduce a new seq2seq variant, called encoder-transformer-decoder. We show that, for all languages considered, there are models reading bytes outperforming the current state-of-the-art word-level baseline. Moreover, the newly introduced encoder-transformer-decoder performs best on the morphologically most involved dataset, Turkish. The large-scale Turkish and Russian machine reading datasets are released to public.

Language Modeling (LM) is a fundamental research topic in a range of areas. Recently, inspired by quantum theory, a novel Quantum Language Model (QLM) has been proposed for Information Retrieval (IR). In this paper, we aim to broaden the theoretical and practical basis of QLM. We develop a Neural Network based Quantum-like Language Model (NNQLM) and apply it to Question Answering. Specifically, based on word embeddings, we design a new density matrix, which represents a sentence (e.g., a question or an answer) and encodes a mixture of semantic subspaces. Such a density matrix, together with a joint representation of the question and the answer, can be integrated into neural network architectures (e.g., 2-dimensional convolutional neural networks). Experiments on the TREC-QA and WIKIQA datasets have verified the effectiveness of our proposed models.

Representing the semantics of words is a long-standing problem for the natural language processing community. Most methods compute word semantics given their textual context in large corpora. More recently, researchers attempted to integrate perceptual and visual features. Most of these works consider the visual appearance of objects to enhance word representations but they ignore the visual environment and context in which objects appear. We propose to unify text-based techniques with vision-based techniques by simultaneously leveraging textual and visual context to learn multimodal word embeddings. We explore various choices for what can serve as a visual context and present an end-to-end method to integrate visual context elements in a multimodal skip-gram model. We provide experiments and extensive analysis of the obtained results.

Semantic dependency graph has been recently proposed as an extension of tree-structured syntactic or semantic representation for natural language sentences. It particularly features the structural property of multi-head, which allows nodes to have multiple heads, resulting in a directed acyclic graph(DAG) parsing problem. Yet most statistical parsers focused exclusively on shallow bi-lexical tree structures, DAG parsing remains under-explored. In this paper, we propose a neural transition-based parser, using a variant of list-based arc-eager transition algorithm for dependency graph parsing. Particularly, two non-trivial improvements are proposed for representing the key components of the transition system, to better capture the semantics of segments and internal sub-graph structures. We test our parser on the SemEval-2016 Task 9 dataset (Chinese) and the SemEval-2015 Task 18 dataset (English). On both benchmark datasets, we obtain superior or comparable results to the best performing systems. Our parser can be further improved with a simple ensemble mechanism, resulting in the state-of-the-art performance.

Generating texts from structured data (e.g., a table) is important for various natural language processing tasks such as question answering and dialog systems. In recent studies, researchers use neural language models and encoder-decoder frameworks for table-to-text generation. However, these neural network-based approaches typically do not model the order of content during text generation. When a human writes a summary based on a given table, he or she would probably consider the content order before wording. In this paper, we propose an order-planning text generation model, where order information is explicitly captured by link-based attention. Then a self-adaptive gate combines the link-based attention with traditional content-based attention. We conducted experiments on the WikiBio dataset and achieve higher performance than previous methods in terms of BLEU, ROUGE, and NIST scores; we also performed ablation tests to analyze each component of our model.

This work performs verb sense induction and clustering based on observed syntactic distributions in a large corpus. VerbNet is a hierarchical clustering of verbs and a useful semantic resource. We address the main drawbacks of VerbNet, by proposing a Bayesian model to build VerbNet-like clusters automatically and with full coverage. Relative to the prior state of the art, we improve accuracy on verb sense induction by over 20% absolute F1. We then propose a new model, inspired by the positive pointwise mutual information (PPMI). Our PPMI-based mixture model permits an extremely efficient sampler, while improving performance. Our best model shows a 4.5% absolute F1 improvement over the best non-PPMI model, with over an order of magnitude less computation time. Though this model is inspired by clustering verb senses, it may be applicable in other situations where multiple items are being sampled as a group.

Copying mechanism shows effectiveness in sequence-to-sequence based neural network models for text generation tasks, such as abstractive sentence summarization and question generation. However, existing works on modeling copying or pointing mechanism only considers single word copying from the source sentences. In this paper, we propose a novel copying framework, named Sequential Copying Networks (SeqCopyNet), which not only learns to copy single words, but also copies sequences from the input sentence. It leverages the pointer networks to explicitly select a sub-span from the source side to target side, and integrates this sequential copying mechanism to the generation process in the encoder-decoder paradigm. Experiments on abstractive sentence summarization and question generation tasks show that the proposed SeqCopyNet can copy meaningful spans and outperforms the baseline models.

This work explores the possibility of learning word order before syntactic concepts such as subject, object, or lexical A hallmark of human word learning is the integration categories or syntactic parse representations are available of cross-situational information even though this information to the learner. It also examines the utility of the acquired is not always reliable as inconsistencies in the wordreferent word order in a joint learner where word order knowledge co-occurrence (e.g., when the referent is absent in a constrains word learning (syntactic bootstrapping) and vice scene or when distracting referents are present) inject noise versa. We propose that the transitional probabilities of the into cross-situational information. It has been suggested thematic roles (in the order of their appearance in the utterance) that bootstrapping cross-situational word learning with the of the referential words (words with action or event learner's belief about the referential intentions of the speaker participant referents) are an invaluable source of information (Frank, Goodman, and Tenenbaum 2009) as well as bootstrapping for learning word order and that they can provide an it with learner's belief about the syntactic regularities initial understanding of the notion of word order in early of language (Yu 2006; Maurits, Perfors, and Navarro stages of language acquisition in the absence of advanced 2009; Alishahi and Fazly 2010; Alishahi and Chrupała 2012; syntactic concepts or representations. We utilize an incremental Abend et al. 2017) allow for disambiguation and should and memory-limited learning algorithm as opposed thus improve word learning. Maurits, Perfors, and Navarro to batch learning algorithms, as we are interested in online (2009) bootstrap word learning with the acquired knowledge learning in embodied agents with computational limitations. of word order in an ideal learner although their model cannot Our model adds the notion of syntax to the word learning

Attention mechanism, including global attention and local attention, plays a key role in neural machine translation (NMT). Global attention attends to all source words for word prediction. In comparison, local attention selectively looks at fixed-window source words. However, alignment weights for the current target word often decrease to the left and right by linear distance centering on the aligned source position and neglect syntax distance constraints. In this paper, we extend the local attention with syntax-distance constraint, which focuses on syntactically related source words with the predicted target word to learning a more effective context vector for predicting translation. Moreover, we further propose a double context NMT architecture, which consists of a global context vector and a syntax-directed context vector from the global attention, to provide more translation performance for NMT from source representation. The experiments on the large-scale Chinese-to-English and English-to-German translation tasks show that the proposed approach achieves a substantial and significant improvement over the baseline system.