ACL 2018 论文列表

Multilingual learning for Neural Named Entity Recognition (NNER) involves jointly training a neural network for multiple languages. Typically, the goal is improving the NER performance of one of the languages (the primary language) using the other assisting languages. We show that the divergence in the tag distributions of the common named entities between the primary and assisting languages can reduce the effectiveness of multilingual learning. To alleviate this problem, we propose a metric based on symmetric KL divergence to filter out the highly divergent training instances in the assisting language. We empirically show that our data selection strategy improves NER performance in many languages, including those with very limited training data.

We propose an entity-centric neural crosslingual coreference model that builds on multi-lingual embeddings and language independent features. We perform both intrinsic and extrinsic evaluations of our model. In the intrinsic evaluation, we show that our model, when trained on English and tested on Chinese and Spanish, achieves competitive results to the models trained directly on Chinese and Spanish respectively. In the extrinsic evaluation, we show that our English model helps achieve superior entity linking accuracy on Chinese and Spanish test sets than the top 2015 TAC system without using any annotated data from Chinese or Spanish.

Recent embedding-based methods in bilingual lexicon induction show good results, but do not take advantage of orthographic features, such as edit distance, which can be helpful for pairs of related languages. This work extends embedding-based methods to incorporate these features, resulting in significant accuracy gains for related languages.

Gender prediction has typically focused on lexical and social network features, yielding good performance, but making systems highly language-, topic-, and platform dependent. Cross-lingual embeddings circumvent some of these limitations, but capture gender-specific style less. We propose an alternative: bleaching text, i.e., transforming lexical strings into more abstract features. This study provides evidence that such features allow for better transfer across languages. Moreover, we present a first study on the ability of humans to perform cross-lingual gender prediction. We find that human predictive power proves similar to that of our bleached models, and both perform better than lexical models.

Attention-based neural machine translation (NMT) models selectively focus on specific source positions to produce a translation, which brings significant improvements over pure encoder-decoder sequence-to-sequence models. This work investigates NMT while replacing the attention component. We study a neural hidden Markov model (HMM) consisting of neural network-based alignment and lexicon models, which are trained jointly using the forward-backward algorithm. We show that the attention component can be effectively replaced by the neural network alignment model and the neural HMM approach is able to provide comparable performance with the state-of-the-art attention-based models on the WMT 2017 German↔English and Chinese→English translation tasks.

In neural machine translation, words are sometimes dropped from the source or generated repeatedly in the translation. We explore novel strategies to address the coverage problem that change only the attention transformation. Our approach allocates fertilities to source words, used to bound the attention each word can receive. We experiment with various sparse and constrained attention transformations and propose a new one, constrained sparsemax, shown to be differentiable and sparse. Empirical evaluation is provided in three languages pairs.

Recent BIO-tagging-based neural semantic role labeling models are very high performing, but assume gold predicates as part of the input and cannot incorporate span-level features. We propose an end-to-end approach for jointly predicting all predicates, arguments spans, and the relations between them. The model makes independent decisions about what relationship, if any, holds between every possible word-span pair, and learns contextualized span representations that provide rich, shared input features for each decision. Experiments demonstrate that this approach sets a new state of the art on PropBank SRL without gold predicates.

Methods for unsupervised hypernym detection may broadly be categorized according to two paradigms: pattern-based and distributional methods. In this paper, we study the performance of both approaches on several hypernymy tasks and find that simple pattern-based methods consistently outperform distributional methods on common benchmark datasets. Our results show that pattern-based models provide important contextual constraints which are not yet captured in distributional methods.

The methods proposed recently for specializing word embeddings according to a particular perspective generally rely on external knowledge. In this article, we propose Pseudofit, a new method for specializing word embeddings according to semantic similarity without any external knowledge. Pseudofit exploits the notion of pseudo-sense for building several representations for each word and uses these representations for making the initial embeddings more generic. We illustrate the interest of Pseudofit for acquiring synonyms and study several variants of Pseudofit according to this perspective.

Verb-noun combinations (VNCs) - e.g., blow the whistle, hit the roof, and see stars - are a common type of English idiom that are ambiguous with literal usages. In this paper we propose and evaluate models for classifying VNC usages as idiomatic or literal, based on a variety of approaches to forming distributed representations. Our results show that a model based on averaging word embeddings performs on par with, or better than, a previously-proposed approach based on skip-thoughts. Idiomatic usages of VNCs are known to exhibit lexico-syntactic fixedness. We further incorporate this information into our models, demonstrating that this rich linguistic knowledge is complementary to the information carried by distributed representations.

To achieve high translation performance, neural machine translation models usually rely on the beam search algorithm for decoding sentences. The beam search finds good candidate translations by considering multiple hypotheses of translations simultaneously. However, as the algorithm produces hypotheses in a monotonic left-to-right order, a hypothesis can not be revisited once it is discarded. We found such monotonicity forces the algorithm to sacrifice some good decoding paths. To mitigate this problem, we relax the monotonic constraint of the beam search by maintaining all found hypotheses in a single priority queue and using a universal score function for hypothesis selection. The proposed algorithm allows discarded hypotheses to be recovered in a later step. Despite its simplicity, we show that the proposed decoding algorithm enhances the quality of selected hypotheses and improve the translations even for high-performance models in English-Japanese translation task.

A sentence can be translated into more than one correct sentences. However, most of the existing neural machine translation models only use one of the correct translations as the targets, and the other correct sentences are punished as the incorrect sentences in the training stage. Since most of the correct translations for one sentence share the similar bag-of-words, it is possible to distinguish the correct translations from the incorrect ones by the bag-of-words. In this paper, we propose an approach that uses both the sentences and the bag-of-words as targets in the training stage, in order to encourage the model to generate the potentially correct sentences that are not appeared in the training set. We evaluate our model on a Chinese-English translation dataset, and experiments show our model outperforms the strong baselines by the BLEU score of 4.55.

We empirically investigate learning from partial feedback in neural machine translation (NMT), when partial feedback is collected by asking users to highlight a correct chunk of a translation. We propose a simple and effective way of utilizing such feedback in NMT training. We demonstrate how the common machine translation problem of domain mismatch between training and deployment can be reduced solely based on chunk-level user feedback. We conduct a series of simulation experiments to test the effectiveness of the proposed method. Our results show that chunk-level feedback outperforms sentence based feedback by up to 2.61% BLEU absolute.

We explore strategies for incorporating target syntax into Neural Machine Translation. We specifically focus on syntax in ensembles containing multiple sentence representations. We formulate beam search over such ensembles using WFSTs, and describe a delayed SGD update training procedure that is especially effective for long representations like linearized syntax. Our approach gives state-of-the-art performance on a difficult Japanese-English task.

Every person speaks or writes their own flavor of their native language, influenced by a number of factors: the content they tend to talk about, their gender, their social status, or their geographical origin. When attempting to perform Machine Translation (MT), these variations have a significant effect on how the system should perform translation, but this is not captured well by standard one-size-fits-all models. In this paper, we propose a simple and parameter-efficient adaptation technique that only requires adapting the bias of the output softmax to each particular user of the MT system, either directly or through a factored approximation. Experiments on TED talks in three languages demonstrate improvements in translation accuracy, and better reflection of speaker traits in the target text.

Neural machine translation (NMT) models are typically trained with fixed-size input and output vocabularies, which creates an important bottleneck on their accuracy and generalization capability. As a solution, various studies proposed segmenting words into sub-word units and performing translation at the sub-lexical level. However, statistical word segmentation methods have recently shown to be prone to morphological errors, which can lead to inaccurate translations. In this paper, we propose to overcome this problem by replacing the source-language embedding layer of NMT with a bi-directional recurrent neural network that generates compositional representations of the input at any desired level of granularity. We test our approach in a low-resource setting with five languages from different morphological typologies, and under different composition assumptions. By training NMT to compose word representations from character n-grams, our approach consistently outperforms (from 1.71 to 2.48 BLEU points) NMT learning embeddings of statistically generated sub-word units.

Traditional Neural machine translation (NMT) involves a fixed training procedure where each sentence is sampled once during each epoch. In reality, some sentences are well-learned during the initial few epochs; however, using this approach, the well-learned sentences would continue to be trained along with those sentences that were not well learned for 10-30 epochs, which results in a wastage of time. Here, we propose an efficient method to dynamically sample the sentences in order to accelerate the NMT training. In this approach, a weight is assigned to each sentence based on the measured difference between the training costs of two iterations. Further, in each epoch, a certain percentage of sentences are dynamically sampled according to their weights. Empirical results based on the NIST Chinese-to-English and the WMT English-to-German tasks show that the proposed method can significantly accelerate the NMT training and improve the NMT performance.

We offer a simple and effective method to seek a better balance between model confidence and length preference for Neural Machine Translation (NMT). Unlike the popular length normalization and coverage models, our model does not require training nor reranking the limited n-best outputs. Moreover, it is robust to large beam sizes, which is not well studied in previous work. On the Chinese-English and English-German translation tasks, our approach yields +0.4 1.5 BLEU improvements over the state-of-the-art baselines.

Effectively using full syntactic parsing information in Neural Networks (NNs) for solving relational tasks, e.g., question similarity, is still an open problem. In this paper, we propose to inject structural representations in NNs by (i) learning a model with Tree Kernels (TKs) on relatively few pairs of questions (few thousands) as gold standard (GS) training data is typically scarce, (ii) predicting labels on a very large corpus of question pairs, and (iii) pre-training NNs on such large corpus. The results on Quora and SemEval question similarity datasets show that NNs using our approach can learn more accurate models, especially after fine tuning on GS.