ACL 2016 论文列表

Context is crucial for identifying argumen-tative relations in text, but many argument mining methods make little use of contextual features. This paper presents context-aware argumentative relation mining that uses features extracted from writing topics as well as from windows of context sentences. Experiments on student essays demonstrate that the proposed features improve predictive performance in two argu-mentative relation classification tasks

We present a new proof that $O_2$ is a multiple context-free language. It contrasts with a recent proof by Salvati (2015) in its avoidance of concepts that seem specific to two-dimensional geometry, such as the complex exponential function. Our simple proof creates realistic prospects of widening the results to higher dimensions. This finding is of central importance to the relation between extreme free word order and classes of grammars used to describe the syntax of natural language.

We present a novel end-to-end neural model to extract entities and relations between them. Our recurrent neural network based model captures both word sequence and dependency tree substructure information by stacking bidirectional tree-structured LSTM-RNNs on bidirectional sequential LSTM-RNNs. This allows our model to jointly represent both entities and relations with shared parameters in a single model. We further encourage detection of entities during training and use of entity information in relation extraction via entity pretraining and scheduled sampling. Our model improves over the state-of-the-art feature-based model on end-to-end relation extraction, achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and ACE2004, respectively. We also show that our LSTM-RNN based model compares favorably to the state-of-the-art CNN based model (in F1-score) on nominal relation classification (SemEval-2010 Task 8). Finally, we present an extensive ablation analysis of several model components.

In this paper, we propose a novel mechanism for enriching the feature vector, for the task of sarcasm detection, with cognitive features extracted from eye-movement patterns of human readers. Sarcasm detection has been a challenging research problem, and its importance for NLP applications such as review summarization, dialog systems and sentiment analysis is well recognized. Sarcasm can often be traced to incongruity that becomes apparent as the full sentence unfolds. This presence of incongruity- implicit or explicit- affects the way readers eyes move through the text. We observe the difference in the behaviour of the eye, while reading sarcastic and non sarcastic sentences. Motivated by this observation, we augment traditional linguistic and stylistic features for sarcasm detection with the cognitive features obtained from readers eye movement data. We perform statistical classification using the enhanced feature set so obtained. The augmented cognitive features improve sarcasm detection by 3.7% (in terms of Fscore), over the performance of the best reported system.

Most machine translation systems construct translations from a closed vocabulary of target word forms, posing problems for translating into languages that have productive compounding processes. We present a simple and effective approach that deals with this problem in two phases. First, we build a classifier that identifies spans of the input text that can be translated into a single compound word in the target language. Then, for each identified span, we generate a pool of possible compounds which are added to the translation model as “synthetic” phrase translations. Experiments reveal that (i) we can effectively predict what spans can be compounded; (ii) our compound generation model produces good compounds; and (iii) modest improvements are possible in end-to-end English‐German and English‐Finnish translation tasks. We additionally introduce KomposEval, a new multi-reference dataset of English phrases and their translations into German compounds.

© 2016 Association for Computational Linguistics. Automatic spoken language assessment systems are becoming increasingly important to meet the demand for English second language learning. This is a challenging task due to the high error rates of, even state-of-the-art, non-native speech recognition. Consequently current systems primarily assess fluency and pronunciation. However, content assessment is essential for full automation. As a first stage it is important to judge whether the speaker responds on topic to test questions designed to elicit spontaneous speech. Standard approaches to off-topic response detection assess similarity between the response and question based on bag-of-words representations. An alternative framework based on Recurrent Neural Network Language Models (RNNLM) is proposed in this paper. The RNNLM is adapted to the topic of each test question. It learns to associate example responses to questions with points in a topic space constructed using these example responses. Classification is done by ranking the topic-conditional posterior probabilities of a response. The RNNLMs associate a broad range of responses with each topic, incorporate sequence information and scale better with additional training data, unlike standard methods. On experiments conducted on data from the Business Language Testing Service (BULATS) this approach outperforms standard approaches.

State-of-the-art sequence labeling systems traditionally require large amounts of task-specific knowledge in the form of hand-crafted features and data pre-processing. In this paper, we introduce a novel neutral network architecture that benefits from both word- and character-level representations automatically, by using combination of bidirectional LSTM, CNN and CRF. Our system is truly end-to-end, requiring no feature engineering or data pre-processing, thus making it applicable to a wide range of sequence labeling tasks. We evaluate our system on two data sets for two sequence labeling tasks --- Penn Treebank WSJ corpus for part-of-speech (POS) tagging and CoNLL 2003 corpus for named entity recognition (NER). We obtain state-of-the-art performance on both the two data --- 97.55\% accuracy for POS tagging and 91.21\% F1 for NER.

Nearly all previous work on neural machine translation (NMT) has used quite restricted vocabularies, perhaps with a subsequent method to patch in unknown words. This paper presents a novel word-character solution to achieving open vocabulary NMT. We build hybrid systems that translate mostly at the word level and consult the character components for rare words. Our character-level recurrent neural networks compute source word representations and recover unknown target words when needed. The twofold advantage of such a hybrid approach is that it is much faster and easier to train than character-based ones; at the same time, it never produces unknown words as in the case of word-based models. On the WMT'15 English to Czech translation task, this hybrid approach offers an addition boost of +2.1-11.4 BLEU points over models that already handle unknown words. Our best system achieves a new state-of-the-art result with 20.7 BLEU score. We demonstrate that our character models can successfully learn to not only generate well-formed words for Czech, a highly-inflected language with a very complex vocabulary, but also build correct representations for English source words.

We construct a humans-in-the-loop supervised learning framework that integrates crowdsourcing feedback and local knowledge to detect job-related tweets from individual and business accounts. Using data-driven ethnography, we examine discourse about work by fusing language-based analysis with temporal, geospational, and labor statistics information.

Recently, there is rising interest in modelling the interactions of text pair with deep neural networks. In this paper, we propose a model of deep fusion LSTMs (DF-LSTMs) to model the strong interaction of text pair in a recursive matching way. Specifically, DF-LSTMs consist of two interdependent LSTMs, each of which models a sequence under the in-fluence of another. We also use exter-nal memory to increase the capacity of LSTMs, thereby possibly capturing more complicated matching patterns. Experiments on two very large datasets demonstrate the efficacy of our proposed architecture. Furthermore, we present an elaborate qualitative analysis of our models, giving an intuitive understanding how our model worked.

We introduce an agreement-based approach to learning parallel lexicons and phrases from non-parallel corpora. The basic idea is to encourage two asymmetric latent-variable translation models (i.e., source-to-target and target-to-source) to agree on identifying latent phrase and word alignments. The agreement is defined at both word and phrase levels. We develop a Viterbi EM algorithm for jointly training the two unidirectional models efficiently. Experiments on the Chinese-English dataset show that agreement-based learning significantly improves both alignment and translation performance.

Automatically recognising medical concepts mentioned in social media messages (e.g. tweets) enables several applications for enhancing health quality of people in a community, e.g. real-time monitoring of infectious diseases in population. However, the discrepancy between the type of language used in social media and medical ontologies poses a major challenge. Existing studies deal with this challenge by employing techniques, such as lexical term matching and statistical machine translation. In this work, we handle the medical concept normalisation at the semantic level. We investigate the use of neural networks to learn the transition between layman’s language used in social media messages and formal medical language used in the descriptions of medical concepts in a standard ontology. We evaluate our approaches using three different datasets, where social media texts are extracted from Twitter messages and blog posts. Our experimental results show that our proposed approaches significantly and consistently outperform existing effective baselines, which achieved state-of-the-art performance on several medical concept normalisation tasks, by up to 44%.

Deep Random Walk (DeepWalk) can learn a latent space representation for describing the topological structure of a network. However, for relational network classification, DeepWalk can be suboptimal as it lacks a mechanism to optimize the objective of the target task. In this paper, we present Discriminative Deep Random Walk (DDRW), a novel method for relational network classification. By solving a joint optimization problem, DDRW can learn the latent space representations that well capture the topological structure and meanwhile are discriminative for the network classification task. Our experimental results on several real social networks demonstrate that DDRW significantly outperforms DeepWalk on multilabel network classification tasks, while retaining the topological structure in the latent space. DDRW is stable and consistently outperforms the baseline methods by various percentages of labeled data. DDRW is also an online method that is scalable and can be naturally parallelized.

We present persona-based models for handling the issue of speaker consistency in neural response generation. A speaker model encodes personas in distributed embeddings that capture individual characteristics such as background information and speaking style. A dyadic speaker-addressee model captures properties of interactions between two interlocutors. Our models yield qualitative performance improvements in both perplexity and BLEU scores over baseline sequence-to-sequence models, with similar gains in speaker consistency as measured by human judges.

Fill-in-the-blank items are commonly featured in computer-assisted language learning (CALL) systems. An item displays a sentence with a blank, and often proposes a number of choices for filling it. These choices should include one correct answer and several plausible distractors. We describe a system that, given an English corpus, automatically generates distractors to produce items for preposition usage. We report a comprehensive evaluation on this system, involving both experts and learners. First, we analyze the diffi-culty levels of machine-generated carrier sentences and distractors, comparing several methods that exploit learner error and learner revision patterns. We show that the quality of machine-generated items approaches that of human-crafted ones. Further, we investigate the extent to which mismatched L1 between the user and the learner corpora affects the quality of distractors. Finally, we measure the system’s impact on the user’s language proficiency in both the short and the long term.

Ekaterina Kochmar’s research is supported by Cambridge English Language Assessment via the ALTA Institute. Ekaterina Shutova’s research is supported by the Leverhulme Trust Early Career Fellowship.

Dialogue topic tracking is a sequential labelling problem of recognizing the topic state at each time step in given dialogue sequences. This paper presents various ar-tificial neural network models for dialogue topic tracking, including convolutional neural networks to account for semantics at each individual utterance, and recurrent neural networks to account for conversational contexts along multiple turns in the dialogue history. The experimental results demonstrate that our proposed models can significantly improve the tracking performances in human-human conversations.

Can we train a system that, on any new input, either says "don't know" or makes a prediction that is guaranteed to be correct? We answer the question in the affirmative provided our model family is well-specified. Specifically, we introduce the unanimity principle: only predict when all models consistent with the training data predict the same output. We operationalize this principle for semantic parsing, the task of mapping utterances to logical forms. We develop a simple, efficient method that reasons over the infinite set of all consistent models by only checking two of the models. We prove that our method obtains 100% precision even with a modest amount of training data from a possibly adversarial distribution. Empirically, we demonstrate the effectiveness of our approach on the standard GeoQuery dataset.

We present the Siamese Continuous Bag of Words (Siamese CBOW) model, a neural network for efficient estimation of high-quality sentence embeddings. Averaging the embeddings of words in a sentence has proven to be a surprisingly successful and efficient way of obtaining sentence embeddings. However, word embeddings trained with the methods currently available are not optimized for the task of sentence representation, and, thus, likely to be suboptimal. Siamese CBOW handles this problem by training word embeddings directly for the purpose of being averaged. The underlying neural network learns word embeddings by predicting, from a sentence representation, its surrounding sentences. We show the robustness of the Siamese CBOW model by evaluating it on 20 datasets stemming from a wide variety of sources.