PyTorchKR Papers

최신 논문

총 5,206편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1910.04659

Multilingual Question Answering from Formatted Text applied to Conversational Agents

Recent advances with language models (e.g. BERT, XLNet, ...), have allowed surpassing human performance on complex NLP tasks such as Reading Comprehension. However, labeled datasets for training are available mostly in English which makes it difficult to acknowledge progress in other languages. Fortunately, models are now pre-trained on unlabeled data from hundreds of languages and exhibit interesting transfer abilities from one language to another. In this paper, we show that multilingual BERT is naturally capable of zero-shot transfer for an extractive Question Answering task (eQA) from English to other languages. More specifically, it outperforms the best previously known baseline for transfer to Japanese and French. Moreover, using a recently published large eQA French dataset, we are able to further show that (1) zero-shot transfer provides results really close to a direct training on the target language and (2) combination of transfer and training on target is the best option overall. We finally present a practical application: a multilingual conversational agent called Kate which answers to HR-related questions in several languages directly from the content of intranet pages.

arXiv:1910.03912

Novel Applications of Factored Neural Machine Translation

In this work, we explore the usefulness of target factors in neural machine translation (NMT) beyond their original purpose of predicting word lemmas and their inflections, as proposed by Garc\`ia-Mart\`inez et al., 2016. For this, we introduce three novel applications of the factored output architecture: In the first one, we use a factor to explicitly predict the word case separately from the target word itself. This allows for information to be shared between different casing variants of a word. In a second task, we use a factor to predict when two consecutive subwords have to be joined, eliminating the need for target subword joining markers. The third task is the prediction of special tokens of the operation sequence NMT model (OSNMT) of Stahlberg et al., 2018. Automatic evaluation on English-to-German and English-to-Turkish tasks showed that integration of such auxiliary prediction tasks into NMT is at least as good as the standard NMT approach. For the OSNMT, we observed a significant improvement in BLEU over the baseline OSNMT implementation due to a reduced output sequence length that resulted from the introduction of the target factors.

arXiv:1910.03475

Named Entity Recognition System for Sindhi Language

Named Entity Recognition (NER) System aims to extract the existing information into the following categories such as: Persons Name, Organization, Location, Date and Time, Term, Designation and Short forms. Now, it is considered to be important aspect for many natural languages processing (NLP) tasks such as: information retrieval system, machine translation system, information extraction system and question answering. Even at a surface level, the understanding of the named entities involved in a document gives richer analytical framework and cross referencing. It has been used for different Arabic Script-Based languages like, Arabic, Persian and Urdu but, Sindhi could not come into being yet. This paper explains the problem of NER in the framework of Sindhi Language and provides relevant solution. The system is developed to tag ten different Named Entities. We have used Ruled based approach for NER system of Sindhi Language. For the training and testing, 936 words were used and calculated performance accuracy of 98.71%.

arXiv:1910.03467

Overcoming the Rare Word Problem for Low-Resource Language Pairs in Neural Machine Translation

Among the six challenges of neural machine translation (NMT) coined by (Koehn and Knowles, 2017), rare-word problem is considered the most severe one, especially in translation of low-resource languages. In this paper, we propose three solutions to address the rare words in neural machine translation systems. First, we enhance source context to predict the target words by connecting directly the source embeddings to the output of the attention component in NMT. Second, we propose an algorithm to learn morphology of unknown words for English in supervised way in order to minimize the adverse effect of rare-word problem. Finally, we exploit synonymous relation from the WordNet to overcome out-of-vocabulary (OOV) problem of NMT. We evaluate our approaches on two low-resource language pairs: English-Vietnamese and Japanese-Vietnamese. In our experiments, we have achieved significant improvements of up to roughly +1.0 BLEU points in both language pairs.

arXiv:1910.03355

An Interactive Machine Translation Framework for Modernizing Historical Documents

Due to the nature of human language, historical documents are hard to comprehend by contemporary people. This limits their accessibility to scholars specialized in the time period in which the documents were written. Modernization aims at breaking this language barrier by generating a new version of a historical document, written in the modern version of the document's original language. However, while it is able to increase the document's comprehension, modernization is still far from producing an error-free version. In this work, we propose a collaborative framework in which a scholar can work together with the machine to generate the new version. We tested our approach on a simulated environment, achieving significant reductions of the human effort needed to produce the modernized version of the document.

arXiv:1910.03320

One-To-Many Multilingual End-to-end Speech Translation

Nowadays, training end-to-end neural models for spoken language translation (SLT) still has to confront with extreme data scarcity conditions. The existing SLT parallel corpora are indeed orders of magnitude smaller than those available for the closely related tasks of automatic speech recognition (ASR) and machine translation (MT), which usually comprise tens of millions of instances. To cope with data paucity, in this paper we explore the effectiveness of transfer learning in end-to-end SLT by presenting a multilingual approach to the task. Multilingual solutions are widely studied in MT and usually rely on ``\textit{target forcing}'', in which multilingual parallel data are combined to train a single model by prepending to the input sequences a language token that specifies the target language. However, when tested in speech translation, our experiments show that MT-like \textit{target forcing}, used as is, is not effective in discriminating among the target languages. Thus, we propose a variant that uses target-language embeddings to shift the input representations in different portions of the space according to the language, so to better support the production of output in the desired target language. Our experiments on end-to-end SLT from English into six languages show important improvements when translating into similar languages, especially when these are supported by scarce data. Further improvements are obtained when using English ASR data as an additional language (up to $+2.5$ BLEU points).

arXiv:1910.03177

Read, Highlight and Summarize: A Hierarchical Neural Semantic Encoder-based Approach

Traditional sequence-to-sequence (seq2seq) models and other variations of the attention-mechanism such as hierarchical attention have been applied to the text summarization problem. Though there is a hierarchy in the way humans use language by forming paragraphs from sentences and sentences from words, hierarchical models have usually not worked that much better than their traditional seq2seq counterparts. This effect is mainly because either the hierarchical attention mechanisms are too sparse using hard attention or noisy using soft attention. In this paper, we propose a method based on extracting the highlights of a document; a key concept that is conveyed in a few sentences. In a typical text summarization dataset consisting of documents that are 800 tokens in length (average), capturing long-term dependencies is very important, e.g., the last sentence can be grouped with the first sentence of a document to form a summary. LSTMs (Long Short-Term Memory) proved useful for machine translation. However, they often fail to capture long-term dependencies while modeling long sequences. To address these issues, we have adapted Neural Semantic Encoders (NSE) to text summarization, a class of memory-augmented neural networks by improving its functionalities and proposed a novel hierarchical NSE that outperforms similar previous models significantly. The quality of summarization was improved by augmenting linguistic factors, namely lemma, and Part-of-Speech (PoS) tags, to each word in the dataset for improved vocabulary coverage and generalization. The hierarchical NSE model on factored dataset outperformed the state-of-the-art by nearly 4 ROUGE points. We further designed and used the first GPU-based self-critical Reinforcement Learning model.

arXiv:1910.03009

Improving Neural Machine Translation Robustness via Data Augmentation: Beyond Back Translation

Neural Machine Translation (NMT) models have been proved strong when translating clean texts, but they are very sensitive to noise in the input. Improving NMT models robustness can be seen as a form of "domain" adaption to noise. The recently created Machine Translation on Noisy Text task corpus provides noisy-clean parallel data for a few language pairs, but this data is very limited in size and diversity. The state-of-the-art approaches are heavily dependent on large volumes of back-translated data. This paper has two main contributions: Firstly, we propose new data augmentation methods to extend limited noisy data and further improve NMT robustness to noise while keeping the models small. Secondly, we explore the effect of utilizing noise from external data in the form of speech transcripts and show that it could help robustness.

arXiv:1910.02238공식 코드 · Papers with Code

How Transformer Revitalizes Character-based Neural Machine Translation: An Investigation on Japanese-Vietnamese Translation Systems

While translating between East Asian languages, many works have discovered clear advantages of using characters as the translation unit. Unfortunately, traditional recurrent neural machine translation systems hinder the practical usage of those character-based systems due to their architectural limitations. They are unfavorable in handling extremely long sequences as well as highly restricted in parallelizing the computations. In this paper, we demonstrate that the new transformer architecture can perform character-based translation better than the recurrent one. We conduct experiments on a low-resource language pair: Japanese-Vietnamese. Our models considerably outperform the state-of-the-art systems which employ word-based recurrent architectures.

arXiv:1910.02211공식 코드 · Papers with Code

On Dimensional Linguistic Properties of the Word Embedding Space

Word embeddings have become a staple of several natural language processing tasks, yet much remains to be understood about their properties. In this work, we analyze word embeddings in terms of their principal components and arrive at a number of novel and counterintuitive observations. In particular, we characterize the utility of variance explained by the principal components as a proxy for downstream performance. Furthermore, through syntactic probing of the principal embedding space, we show that the syntactic information captured by a principal component does not correlate with the amount of variance it explains. Consequently, we investigate the limitations of variance based embedding post-processing and demonstrate that such post-processing is counter-productive in sentence classification and machine translation tasks. Finally, we offer a few precautionary guidelines on applying variance based embedding post-processing and explain why non-isotropic geometry might be integral to word embedding performance.

arXiv:1910.02612공식 코드 · Papers with Code

Compositional Generalization for Primitive Substitutions

Compositional generalization is a basic mechanism in human language learning, but current neural networks lack such ability. In this paper, we conduct fundamental research for encoding compositionality in neural networks. Conventional methods use a single representation for the input sentence, making it hard to apply prior knowledge of compositionality. In contrast, our approach leverages such knowledge with two representations, one generating attention maps, and the other mapping attended input words to output symbols. We reduce the entropy in each representation to improve generalization. Our experiments demonstrate significant improvements over the conventional methods in five NLP tasks including instruction learning and machine translation. In the SCAN domain, it boosts accuracies from 14.0% to 98.8% in Jump task, and from 92.0% to 99.7% in TurnLeft task. It also beats human performance on a few-shot learning task. We hope the proposed approach can help ease future research towards human-level compositional language learning.

arXiv:1910.02602

Human Action Sequence Classification

This paper classifies human action sequences from videos using a machine translation model. In contrast to classical human action classification which outputs a set of actions, our method output a sequence of action in the chronological order of the actions performed by the human. Therefore our method is evaluated using sequential performance measures such as Bilingual Evaluation Understudy (BLEU) scores. Action sequence classification has many applications such as learning from demonstration, action segmentation, detection, localization and video captioning. Furthermore, we use our model that is trained to output action sequences to solve downstream tasks; such as video captioning and action localization. We obtain state of the art results for video captioning in challenging Charades dataset obtaining BLEU-4 score of 34.8 and METEOR score of 33.6 outperforming previous state-of-the-art of 18.8 and 19.5 respectively. Similarly, on ActivityNet captioning, we obtain excellent results in-terms of ROUGE (20.24) and CIDER (37.58) scores. For action localization, without using any explicit start/end action annotations, our method obtains localization performance of 22.2 mAP outperforming prior fully supervised methods.

arXiv:1910.02555공식 코드 · Papers with Code

Domain Differential Adaptation for Neural Machine Translation

Neural networks are known to be data hungry and domain sensitive, but it is nearly impossible to obtain large quantities of labeled data for every domain we are interested in. This necessitates the use of domain adaptation strategies. One common strategy encourages generalization by aligning the global distribution statistics between source and target domains, but one drawback is that the statistics of different domains or tasks are inherently divergent, and smoothing over these differences can lead to sub-optimal performance. In this paper, we propose the framework of {\it Domain Differential Adaptation (DDA)}, where instead of smoothing over these differences we embrace them, directly modeling the difference between domains using models in a related task. We then use these learned domain differentials to adapt models for the target task accordingly. Experimental results on domain adaptation for neural machine translation demonstrate the effectiveness of this strategy, achieving consistent improvements over other alternative adaptation strategies in multiple experimental settings.

arXiv:1910.02766

Adversarial reconstruction for Multi-modal Machine Translation

Even with the growing interest in problems at the intersection of Computer Vision and Natural Language, grounding (i.e. identifying) the components of a structured description in an image still remains a challenging task. This contribution aims to propose a model which learns grounding by reconstructing the visual features for the Multi-modal translation task. Previous works have partially investigated standard approaches such as regression methods to approximate the reconstruction of a visual input. In this paper, we propose a different and novel approach which learns grounding by adversarial feedback. To do so, we modulate our network following the recent promising adversarial architectures and evaluate how the adversarial response from a visual reconstruction as an auxiliary task helps the model in its learning. We report the highest scores in term of BLEU and METEOR metrics on the different datasets.

arXiv:1910.02754

On Leveraging the Visual Modality for Neural Machine Translation

Leveraging the visual modality effectively for Neural Machine Translation (NMT) remains an open problem in computational linguistics. Recently, Caglayan et al. posit that the observed gains are limited mainly due to the very simple, short, repetitive sentences of the Multi30k dataset (the only multimodal MT dataset available at the time), which renders the source text sufficient for context. In this work, we further investigate this hypothesis on a new large scale multimodal Machine Translation (MMT) dataset, How2, which has 1.57 times longer mean sentence length than Multi30k and no repetition. We propose and evaluate three novel fusion techniques, each of which is designed to ensure the utilization of visual context at different stages of the Sequence-to-Sequence transduction pipeline, even under full linguistic context. However, we still obtain only marginal gains under full linguistic context and posit that visual embeddings extracted from deep vision models (ResNet for Multi30k, ResNext for How2) do not lend themselves to increasing the discriminativeness between the vocabulary elements at token level prediction in NMT. We demonstrate this qualitatively by analyzing attention distribution and quantitatively through Principal Component Analysis, arriving at the conclusion that it is the quality of the visual embeddings rather than the length of sentences, which need to be improved in existing MMT datasets.

arXiv:1910.01859

Modeling Confidence in Sequence-to-Sequence Models

Recently, significant improvements have been achieved in various natural language processing tasks using neural sequence-to-sequence models. While aiming for the best generation quality is important, ultimately it is also necessary to develop models that can assess the quality of their output. In this work, we propose to use the similarity between training and test conditions as a measure for models' confidence. We investigate methods solely using the similarity as well as methods combining it with the posterior probability. While traditionally only target tokens are annotated with confidence measures, we also investigate methods to annotate source tokens with confidence. By learning an internal alignment model, we can significantly improve confidence projection over using state-of-the-art external alignment tools. We evaluate the proposed methods on downstream confidence estimation for machine translation (MT). We show improvements on segment-level confidence estimation as well as on confidence estimation for source tokens. In addition, we show that the same methods can also be applied to other tasks using sequence-to-sequence models. On the automatic speech recognition (ASR) task, we are able to find 60% of the errors by looking at 20% of the data.

arXiv:1910.01289

Neural Zero-Inflated Quality Estimation Model For Automatic Speech Recognition System

The performances of automatic speech recognition (ASR) systems are usually evaluated by the metric word error rate (WER) when the manually transcribed data are provided, which are, however, expensively available in the real scenario. In addition, the empirical distribution of WER for most ASR systems usually tends to put a significant mass near zero, making it difficult to simulate with a single continuous distribution. In order to address the two issues of ASR quality estimation (QE), we propose a novel neural zero-inflated model to predict the WER of the ASR result without transcripts. We design a neural zero-inflated beta regression on top of a bidirectional transformer language model conditional on speech features (speech-BERT). We adopt the pre-training strategy of token level mask language modeling for speech-BERT as well, and further fine-tune with our zero-inflated layer for the mixture of discrete and continuous outputs. The experimental results show that our approach achieves better performance on WER prediction in the metrics of Pearson and MAE, compared with most existed quality estimation algorithms for ASR or machine translation.

arXiv:1910.00643공식 코드 · Papers with Code

SlowMo: Improving Communication-Efficient Distributed SGD with Slow Momentum

Distributed optimization is essential for training large models on large datasets. Multiple approaches have been proposed to reduce the communication overhead in distributed training, such as synchronizing only after performing multiple local SGD steps, and decentralized methods (e.g., using gossip algorithms) to decouple communications among workers. Although these methods run faster than AllReduce-based methods, which use blocking communication before every update, the resulting models may be less accurate after the same number of updates. Inspired by the BMUF method of Chen & Huo (2016), we propose a slow momentum (SlowMo) framework, where workers periodically synchronize and perform a momentum update, after multiple iterations of a base optimization algorithm. Experiments on image classification and machine translation tasks demonstrate that SlowMo consistently yields improvements in optimization and generalization performance relative to the base optimizer, even when the additional overhead is amortized over many updates so that the SlowMo runtime is on par with that of the base optimizer. We provide theoretical convergence guarantees showing that SlowMo converges to a stationary point of smooth non-convex losses. Since BMUF can be expressed through the SlowMo framework, our results also correspond to the first theoretical convergence guarantees for BMUF.

arXiv:1910.00373

A Survey of Methods to Leverage Monolingual Data in Low-resource Neural Machine Translation

Neural machine translation has become the state-of-the-art for language pairs with large parallel corpora. However, the quality of machine translation for low-resource languages leaves much to be desired. There are several approaches to mitigate this problem, such as transfer learning, semi-supervised and unsupervised learning techniques. In this paper, we review the existing methods, where the main idea is to exploit the power of monolingual data, which, compared to parallel, is usually easier to obtain and significantly greater in amount.

arXiv:1910.00368

Application of Low-resource Machine Translation Techniques to Russian-Tatar Language Pair

Neural machine translation is the current state-of-the-art in machine translation. Although it is successful in a resource-rich setting, its applicability for low-resource language pairs is still debatable. In this paper, we explore the effect of different techniques to improve machine translation quality when a parallel corpus is as small as 324 000 sentences, taking as an example previously unexplored Russian-Tatar language pair. We apply such techniques as transfer learning and semi-supervised learning to the base Transformer model, and empirically show that the resulting models improve Russian to Tatar and Tatar to Russian translation quality by +2.57 and +3.66 BLEU, respectively.