NAACL 2016 论文列表

Given a pair of sentences, we present computational models to assess if one sentence is simpler to read than the other. While existing models explored the usage of phrase structure features using a non-incremental parser, experimental evidence suggests that the human language processor works incrementally. We empirically evaluate if syntactic features from an incremental CCG parser are more useful than features from a non-incremental phrase structure parser. Our evaluation on Simple and Standard Wikipedia sentence pairs suggests that incremental CCG features are indeed more useful than phrase structure features achieving 0.44 points gain in performance. Incremental CCG parser also gives significant improvements in speed (12 times faster) in comparison to the phrase structure parser. Furthermore, with the addition of psycholinguistic features, we achieve the strongest result to date reported on this task. Our code and data can be downloaded from https://github. com/bharatambati/sent-compl.

,

We provide the first extensive evaluation of how using different types of context to learn skip-gram word embeddings affects performance on a wide range of intrinsic and extrinsic NLP tasks. Our results suggest that while intrinsic tasks tend to exhibit a clear preference to particular types of contexts and higher dimensionality, more careful tuning is required for finding the optimal settings for most of the extrinsic tasks that we considered. Furthermore, for these extrinsic tasks, we find that once the benefit from increasing the embedding dimensionality is mostly exhausted, simple concatenation of word embeddings, learned with different context types, can yield further performance gains. As an additional contribution, we propose a new variant of the skip-gram model that learns word embeddings from weighted contexts of substitute words.

Modern NLP models rely heavily on engineered features, which often combine word and contextual information into complex lexical features. Such combination results in large numbers of features, which can lead to over-fitting. We present a new model that represents complex lexical features---comprised of parts for words, contextual information and labels---in a tensor that captures conjunction information among these parts. We apply low-rank tensor approximations to the corresponding parameter tensors to reduce the parameter space and improve prediction speed. Furthermore, we investigate two methods for handling features that include $n$-grams of mixed lengths. Our model achieves state-of-the-art results on tasks in relation extraction, PP-attachment, and preposition disambiguation.

Coreference resolution is one of the first stages in deep language understanding and its importance has been well recognized in the natural language processing community. In this paper, we propose a generative, unsupervised ranking model for entity coreference resolution by introducing resolution mode variables. Our unsupervised system achieves 58.44% F1 score of the CoNLL metric on the English data from the CoNLL-2012 shared task (Pradhan et al., 2012), outperforming the Stanford deterministic system (Lee et al., 2013) by 3.01%.

There is a significant gap between the performance of a coreference resolution system on gold mentions and on system mentions. This gap is due to the large and unbalanced search space in coreference resolution when using system mentions. In this paper we show that search space pruning is a simple but effi-cient way of improving coreference resolvers. By incorporating our pruning method in one of the state-of-the-art coreference resolution systems, we achieve the best reported over-all score on the CoNLL 2012 English test set. A version of our pruning method is available with the Cort coreference resolution source code.

There is compelling evidence that coreference prediction would benefit from modeling global information about entity-clusters. Yet, state-of-the-art performance can be achieved with systems treating each mention prediction independently, which we attribute to the inherent difficulty of crafting informative cluster-level features. We instead propose to use recurrent neural networks (RNNs) to learn latent, global representations of entity clusters directly from their mentions. We show that such representations are especially useful for the prediction of pronominal mentions, and can be incorporated into an end-to-end coreference system that outperforms the state of the art without requiring any additional search.

Dropped Pronouns (DP) in which pronouns are frequently dropped in the source language but should be retained in the target language are challenge in machine translation. In response to this problem, we propose a semisupervised approach to recall possibly missing pronouns in the translation. Firstly, we build training data for DP generation in which the DPs are automatically labelled according to the alignment information from a parallel corpus. Secondly, we build a deep learning-based DP generator for input sentences in decoding when no corresponding references exist. More specifically, the generation is two-phase: (1) DP position detection, which is modeled as a sequential labelling task with recurrent neural networks; and (2) DP prediction, which employs a multilayer perceptron with rich features. Finally, we integrate the above outputs into our translation system to recall missing pronouns by both extracting rules from the DP-labelled training data and translating the DP-generated input sentences. Experimental results show that our approach achieves a significant improvement of 1.58 BLEU points in translation performance with 66% F-score for DP generation accuracy.

Artificial neural networks are powerful models, which have been widely applied into many aspects of machine translation, such as language modeling and translation modeling. Though notable improvements have been made in these areas, the reordering problem still remains a challenge in statistical machine translations. In this paper, we present a novel neural reordering model that directly models word pairs and alignment. By utilizing LSTM recurrent neural networks, much longer context could be learned for reordering prediction. Experimental results on NIST OpenMT12 Arabic-English and Chinese-English 1000-best rescoring task show that our LSTM neural reordering feature is robust and achieves significant improvements over various baseline systems.

Computational approaches to simultaneous interpretation are stymied by how little we know about the tactics human interpreters use. We produce a parallel corpus of translated and simultaneously interpreted text and study differences between them through a computational approach. Our analysis reveals that human interpreters regularly apply several effective tactics to reduce translation latency, including sentence segmentation and passivization. In addition to these unique, clever strategies, we show that limited human memory also causes other idiosyncratic properties of human interpretation such as generalization and omission of source content.

This paper introduces information density and machine translation quality estimation inspired features to automatically detect and classify human translated texts. We investigate two settings: discriminating between translations and comparable originally authored texts, and distinguishing two levels of translation professionalism. Our framework is based on delexicalised sentence-level dense feature vector representations combined with a supervised machine learning approach. The results show state-of-the-art performance for mixed-domain translationese detection with information density and quality estimation based features, while results on translation expertise classification are mixed.

For many low-resource languages, spoken language resources are more likely to be annotated with translations than transcriptions. This bilingual speech data can be used for word-spotting, spoken document retrieval, and even for documentation of endangered languages. We experiment with the neural, attentional model applied to this data. On phoneto-word alignment and translation reranking tasks, we achieve large improvements relative to several baselines. On the more challenging speech-to-word alignment task, our model nearly matches GIZA++’s performance on gold transcriptions, but without recourse to transcriptions or to a lexicon.

Textual similarity measurement is a challenging problem, as it requires understanding the semantics of input sentences. Most previous neural network models use coarse-grained sentence modeling, which has difficulty capturing fine-grained word-level information for semantic comparisons. As an alternative, we propose to explicitly model pairwise word interactions and present a novel similarity focus mechanism to identify important correspondences for better similarity measurement. Our ideas are implemented in a novel neural network architecture that demonstrates state-ofthe-art accuracy on three SemEval tasks and two answer selection tasks.

Identification of short text similarity ( STS ) is a high-utility NLP task with applications in a variety of domains. We explore adaptation of STS algorithms to different target domains and applications. A two-level hierarchical Bayesian model is employed for domain adaptation ( DA ) of a linear STS model to text from different sources (e.g., news, tweets). This model is then further extended for multitask learning ( MTL ) of three related tasks: STS , short answer scoring ( SAS ) and answer sentence ranking ( ASR ). In our experiments, the adaptive model demonstrates better overall cross-domain and cross-task performance over two non-adaptive baselines.

Recurrent neural networks, particularly long short-term memory (LSTM), have recently shown to be very effective in a wide range of sequence modeling problems, core to which is effective learning of distributed representation for subsequences as well as the sequences they form. An assumption in almost all the previous models, however, posits that the learned representation (e.g., a distributed representation for a sentence), is fully compositional from the atomic components (e.g., representations for words), while non-compositionality is a basic phenomenon in human languages. In this paper, we relieve the assumption by extending the chain-structured LSTM to directed acyclic graphs (DAGs), with the aim to endow linear-chain LSTMs with the capability of considering compositionality together with non-compositionality in the same semantic composition framework. From a more general viewpoint, the proposed models incorporate additional prior knowledge into recurrent neural networks, which is interesting to us, considering most NLP tasks have relatively small training data and appropriate prior knowledge could be beneficial to help cover missing semantics. Our experiments on sentiment composition demonstrate that the proposed models achieve the state-of-the-art performance, outperforming models that lack

This paper proposes a novel translation-based knowledge graph embedding that preserves the logical properties of relations such as transitivity and symmetricity. The embedding space generated by existing translation-based embeddings do not represent transitive and symmetric relations precisely, because they ignore the role of entities in triples. Thus, we introduce a role-specific projection which maps an entity to distinct vectors according to its role in a triple. That is, a head entity is projected onto an embedding space by a head projection operator, and a tail entity is projected by a tail projection operator. This idea is applied to TransE, TransR, and TransD to produce lppTransE, lppTransR, and lppTransD, respectively. According to the experimental results on link prediction and triple classification, the proposed logical property preserving embeddings show the state-of-the-art performance at both tasks. These results prove that it is critical to preserve logical properties of relations while embedding knowledge graphs, and the proposed method does it effectively.

Can crowdsourced annotation of training data boost performance for relation extraction over methods based solely on distant supervision? While crowdsourcing has been shown effective for many NLP tasks, previous researchers found only minimal improvement when applying the method to relation extraction. This paper demonstrates that a much larger boost is possible, e.g., raising F1 from 0.40 to 0.60. Furthermore, the gains are due to a simple, generalizable technique, Gated Instruction , which combines an interactive tutorial, feedback to correct errors during training, and improved screening.

Universal schema builds a knowledge base (KB) of entities and relations by jointly embedding all relation types from input KBs as well as textual patterns expressing relations from raw text. In most previous applications of universal schema, each textual pattern is represented as a single embedding, preventing generalization to unseen patterns. Recent work employs a neural network to capture patterns' compositional semantics, providing generalization to all possible input text. In response, this paper introduces significant further improvements to the coverage and flexibility of universal schema relation extraction: predictions for entities unseen in training and multilingual transfer learning to domains with no annotation. We evaluate our model through extensive experiments on the English and Spanish TAC KBP benchmark, outperforming the top system from TAC 2013 slot-filling using no handwritten patterns or additional annotation. We also consider a multilingual setting in which English training data entities overlap with the seed KB, but Spanish text does not. Despite having no annotation for Spanish data, we train an accurate predictor, with additional improvements obtained by tying word embeddings across languages. Furthermore, we find that multilingual training improves English relation extraction accuracy. Our approach is thus suited to broad-coverage automated knowledge base construction in a variety of languages and domains.

Neural encoder-decoder models of machine translation have achieved impressive results, rivalling traditional translation models. However their modelling formulation is overly simplistic, and omits several key inductive biases built into traditional models. In this paper we extend the attentional neural translation model to include structural biases from word based alignment models, including positional bias, Markov conditioning, fertility and agreement over translation directions. We show improvements over a baseline attentional model and standard phrase-based model over several language pairs, evaluating on difficult languages in a low resource setting.