EMNLP 2019 论文列表

We explore whether it is possible to leverage eye-tracking data in an RNN dependency parser (for English) when such information is only available during training - i.e. no aggregated or token-level gaze features are used at inference time. To do so, we train a multitask learning model that parses sentences as sequence labeling and leverages gaze features as auxiliary tasks. Our method also learns to train from disjoint datasets, i.e. it can be used to test whether already collected gaze features are useful to improve the performance on new non-gazed annotated treebanks. Accuracy gains are modest but positive, showing the feasibility of the approach. It can serve as a first step towards architectures that can better leverage eye-tracking data or other complementary information available only for training sentences, possibly leading to improvements in syntactic parsing.

Core to the vision-and-language navigation (VLN) challenge is building robust instruction representations and action decoding schemes, which can generalize well to previously unseen instructions and environments. In this paper, we report two simple but highly effective methods to address these challenges and lead to a new state-of-the-art performance. First, we adapt large-scale pretrained language models to learn text representations that generalize better to previously unseen instructions. Second, we propose a stochastic sampling scheme to reduce the considerable gap between the expert actions in training and sampled actions in test, so that the agent can learn to correct its own mistakes during long sequential action decoding. Combining the two techniques, we achieve a new state of the art on the Room-to-Room benchmark with 6% absolute gain over the previous best result (47% -> 53%) on the Success Rate weighted by Path Length metric.

Grounding is crucial for natural language understanding. An important subtask is to understand modified color expressions, such as “light blue”. We present a model of color modifiers that, compared with previous additive models in RGB space, learns more complex transformations. In addition, we present a model that operates in the HSV color space. We show that certain adjectives are better modeled in that space. To account for all modifiers, we train a hard ensemble model that selects a color space depending on the modifier-color pair. Experimental results show significant and consistent improvements compared to the state-of-the-art baseline model.

We propose weakly supervised language localization networks (WSLLN) to detect events in long, untrimmed videos given language queries. To learn the correspondence between visual segments and texts, most previous methods require temporal coordinates (start and end times) of events for training, which leads to high costs of annotation. WSLLN relieves the annotation burden by training with only video-sentence pairs without accessing to temporal locations of events. With a simple end-to-end structure, WSLLN measures segment-text consistency and conducts segment selection (conditioned on the text) simultaneously. Results from both are merged and optimized as a video-sentence matching problem. Experiments on ActivityNet Captions and DiDeMo demonstrate that WSLLN achieves state-of-the-art performance.

Popular metrics used for evaluating image captioning systems, such as BLEU and CIDEr, provide a single score to gauge the system’s overall effectiveness. This score is often not informative enough to indicate what specific errors are made by a given system. In this study, we present a fine-grained evaluation method REO for automatically measuring the performance of image captioning systems. REO assesses the quality of captions from three perspectives: 1) Relevance to the ground truth, 2) Extraness of the content that is irrelevant to the ground truth, and 3) Omission of the elements in the images and human references. Experiments on three benchmark datasets demonstrate that our method achieves a higher consistency with human judgments and provides more intuitive evaluation results than alternative metrics.

Object detection plays an important role in current solutions to vision and language tasks like image captioning and visual question answering. However, popular models like Faster R-CNN rely on a costly process of annotating ground-truths for both the bounding boxes and their corresponding semantic labels, making it less amenable as a primitive task for transfer learning. In this paper, we examine the effect of decoupling box proposal and featurization for down-stream tasks. The key insight is that this allows us to leverage a large amount of labeled annotations that were previously unavailable for standard object detection benchmarks. Empirically, we demonstrate that this leads to effective transfer learning and improved image captioning and visual question answering models, as measured on publicly-available benchmarks.

With the aim of promoting and understanding the multilingual version of image search, we leverage visual object detection and propose a model with diverse multi-head attention to learn grounded multilingual multimodal representations. Specifically, our model attends to different types of textual semantics in two languages and visual objects for fine-grained alignments between sentences and images. We introduce a new objective function which explicitly encourages attention diversity to learn an improved visual-semantic embedding space. We evaluate our model in the German-Image and English-Image matching tasks on the Multi30K dataset, and in the Semantic Textual Similarity task with the English descriptions of visual content. Results show that our model yields a significant performance gain over other methods in all of the three tasks.

A typical cross-lingual transfer learning approach boosting model performance on a language is to pre-train the model on all available supervised data from another language. However, in large-scale systems this leads to high training times and computational requirements. In addition, characteristic differences between the source and target languages raise a natural question of whether source data selection can improve the knowledge transfer. In this paper, we address this question and propose a simple but effective language model based source-language data selection method for cross-lingual transfer learning in large-scale spoken language understanding. The experimental results show that with data selection i) source data and hence training speed is reduced significantly and ii) model performance is improved.

Prior work on training generative Visual Dialog models with reinforcement learning ((Das et al., ICCV 2017) has explored a Q-Bot-A-Bot image-guessing game and shown that this ‘self-talk’ approach can lead to improved performance at the downstream dialog-conditioned image-guessing task. However, this improvement saturates and starts degrading after a few rounds of interaction, and does not lead to a better Visual Dialog model. We find that this is due in part to repeated interactions between Q-Bot and A-BOT during self-talk, which are not informative with respect to the image. To improve this, we devise a simple auxiliary objective that incentivizes Q-Bot to ask diverse questions, thus reducing repetitions and in turn enabling A-Bot to explore a larger state space during RL i.e. be exposed to more visual concepts to talk about, and varied questions to answer. We evaluate our approach via a host of automatic metrics and human studies, and demonstrate that it leads to better dialog, i.e. dialog that is more diverse (i.e. less repetitive), consistent (i.e. has fewer conflicting exchanges), fluent (i.e., more human-like), and detailed, while still being comparably image-relevant as prior work and ablations.

Diacritic restoration has gained importance with the growing need for machines to understand written texts. The task is typically modeled as a sequence labeling problem and currently Bidirectional Long Short Term Memory (BiLSTM) models provide state-of-the-art results. Recently, Bai et al. (2018) show the advantages of Temporal Convolutional Neural Networks (TCN) over Recurrent Neural Networks (RNN) for sequence modeling in terms of performance and computational resources. As diacritic restoration benefits from both previous as well as subsequent timesteps, we further apply and evaluate a variant of TCN, Acausal TCN (A-TCN), which incorporates context from both directions (previous and future) rather than strictly incorporating previous context as in the case of TCN. A-TCN yields significant improvement over TCN for diacritization in three different languages: Arabic, Yoruba, and Vietnamese. Furthermore, A-TCN and BiLSTM have comparable performance, making A-TCN an efficient alternative over BiLSTM since convolutions can be trained in parallel. A-TCN is significantly faster than BiLSTM at inference time (270% 334% improvement in the amount of text diacritized per minute).

Korean morphological analysis has been considered as a sequence of morpheme processing and POS tagging. Thus, a pipeline model of the tasks has been adopted widely by previous studies. However, the model has a problem that it cannot utilize interactions among the tasks. This paper formulates Korean morphological analysis as a combination of the tasks and presents a tied sequence-to-sequence multi-task model for training the two tasks simultaneously without any explicit regularization. The experiments prove the proposed model achieves the state-of-the-art performance.

Neural machine translation (NMT) has achieved new state-of-the-art performance in translating ambiguous words. However, it is still unclear which component dominates the process of disambiguation. In this paper, we explore the ability of NMT encoders and decoders to disambiguate word senses by evaluating hidden states and investigating the distributions of self-attention. We train a classifier to predict whether a translation is correct given the representation of an ambiguous noun. We find that encoder hidden states outperform word embeddings significantly which indicates that encoders adequately encode relevant information for disambiguation into hidden states. In contrast to encoders, the effect of decoder is different in models with different architectures. Moreover, the attention weights and attention entropy show that self-attention can detect ambiguous nouns and distribute more attention to the context.

Grammar induction aims to discover syntactic structures from unannotated sentences. In this paper, we propose a framework in which the learning process of the grammar model of one language is influenced by knowledge from the model of another language. Unlike previous work on multilingual grammar induction, our approach does not rely on any external resource, such as parallel corpora, word alignments or linguistic phylogenetic trees. We propose three regularization methods that encourage similarity between model parameters, dependency edge scores, and parse trees respectively. We deploy our methods on a state-of-the-art unsupervised discriminative parser and evaluate it on both transfer grammar induction and bilingual grammar induction. Empirical results on multiple languages show that our methods outperform strong baselines.

The recent success of neural machine translation models relies on the availability of high quality, in-domain data. Domain adaptation is required when domain-specific data is scarce or nonexistent. Previous unsupervised domain adaptation strategies include training the model with in-domain copied monolingual or back-translated data. However, these methods use generic representations for text regardless of domain shift, which makes it infeasible for translation models to control outputs conditional on a specific domain. In this work, we propose an approach that adapts models with domain-aware feature embeddings, which are learned via an auxiliary language modeling task. Our approach allows the model to assign domain-specific representations to words and output sentences in the desired domain. Our empirical results demonstrate the effectiveness of the proposed strategy, achieving consistent improvements in multiple experimental settings. In addition, we show that combining our method with back translation can further improve the performance of the model.

This paper highlights the impressive utility of multi-parallel corpora for transfer learning in a one-to-many low-resource neural machine translation (NMT) setting. We report on a systematic comparison of multistage fine-tuning configurations, consisting of (1) pre-training on an external large (209k–440k) parallel corpus for English and a helping target language, (2) mixed pre-training or fine-tuning on a mixture of the external and low-resource (18k) target parallel corpora, and (3) pure fine-tuning on the target parallel corpora. Our experiments confirm that multi-parallel corpora are extremely useful despite their scarcity and content-wise redundancy thus exhibiting the true power of multilingualism. Even when the helping target language is not one of the target languages of our concern, our multistage fine-tuning can give 3–9 BLEU score gains over a simple one-to-one model.

Although self-attention networks (SANs) have advanced the state-of-the-art on various NLP tasks, one criticism of SANs is their ability of encoding positions of input words (Shaw et al., 2018). In this work, we propose to augment SANs with structural position representations to model the latent structure of the input sentence, which is complementary to the standard sequential positional representations. Specifically, we use dependency tree to represent the grammatical structure of a sentence, and propose two strategies to encode the positional relationships among words in the dependency tree. Experimental results on NIST Chinese-to-English and WMT14 English-to-German translation tasks show that the proposed approach consistently boosts performance over both the absolute and relative sequential position representations.

Beam search is universally used in (full-sentence) machine translation but its application to simultaneous translation remains highly non-trivial, where output words are committed on the fly. In particular, the recently proposed wait-k policy (Ma et al., 2018) is a simple and effective method that (after an initial wait) commits one output word on receiving each input word, making beam search seemingly inapplicable. To address this challenge, we propose a new speculative beam search algorithm that hallucinates several steps into the future in order to reach a more accurate decision by implicitly benefiting from a target language model. This idea makes beam search applicable for the first time to the generation of a single word in each step. Experiments over diverse language pairs show large improvement compared to previous work.

Despite impressive empirical successes of neural machine translation (NMT) on standard benchmarks, limited parallel data impedes the application of NMT models to many language pairs. Data augmentation methods such as back-translation make it possible to use monolingual data to help alleviate these issues, but back-translation itself fails in extreme low-resource scenarios, especially for syntactically divergent languages. In this paper, we propose a simple yet effective solution, whereby target-language sentences are re-ordered to match the order of the source and used as an additional source of training-time supervision. Experiments with simulated low-resource Japanese-to-English, and real low-resource Uyghur-to-English scenarios find significant improvements over other semi-supervised alternatives.

Bilingual lexicons are valuable resources used by professional human translators. While these resources can be easily incorporated in statistical machine translation, it is unclear how to best do so in the neural framework. In this work, we present the HABLex dataset, designed to test methods for bilingual lexicon integration into neural machine translation. Our data consists of human generated alignments of words and phrases in machine translation test sets in three language pairs (Russian-English, Chinese-English, and Korean-English), resulting in clean bilingual lexicons which are well matched to the reference. We also present two simple baselines - constrained decoding and continued training - and an improvement to continued training to address overfitting.