PyTorchKR Papers

최신 논문

총 5,206편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1911.03864

Improving Transformer Models by Reordering their Sublayers

Multilayer transformer networks consist of interleaved self-attention and feedforward sublayers. Could ordering the sublayers in a different pattern lead to better performance? We generate randomly ordered transformers and train them with the language modeling objective. We observe that some of these models are able to achieve better performance than the interleaved baseline, and that those successful variants tend to have more self-attention at the bottom and more feedforward sublayers at the top. We propose a new transformer pattern that adheres to this property, the sandwich transformer, and show that it improves perplexity on multiple word-level and character-level language modeling benchmarks, at no cost in parameters, memory, or training time. However, the sandwich reordering pattern does not guarantee performance gains across every task, as we demonstrate on machine translation models. Instead, we suggest that further exploration of task-specific sublayer reorderings is needed in order to unlock additional gains.

arXiv:1911.03853

Modelling Bahdanau Attention using Election methods aided by Q-Learning

Neural Machine Translation has lately gained a lot of "attention" with the advent of more and more sophisticated but drastically improved models. Attention mechanism has proved to be a boon in this direction by providing weights to the input words, making it easy for the decoder to identify words representing the present context. But by and by, as newer attention models with more complexity came into development, they involved large computation, making inference slow. In this paper, we have modelled the attention network using techniques resonating with social choice theory. Along with that, the attention mechanism, being a Markov Decision Process, has been represented by reinforcement learning techniques. Thus, we propose to use an election method ($k$-Borda), fine-tuned using Q-learning, as a replacement for attention networks. The inference time for this network is less than a standard Bahdanau translator, and the results of the translation are comparable. This not only experimentally verifies the claims stated above but also helped provide a faster inference.

arXiv:1911.04292

Diversity by Phonetics and its Application in Neural Machine Translation

We introduce a powerful approach for Neural Machine Translation (NMT), whereby, during training and testing, together with the input we provide its phonetic encoding and the variants of such an encoding. This way we obtain very significant improvements up to 4 BLEU points over the state-of-the-art large-scale system. The phonetic encoding is the first part of our contribution, with a second being a theory that aims to understand the reason for this improvement. Our hypothesis states that the phonetic encoding helps NMT because it encodes a procedure to emphasize the difference between semantically diverse sentences. We conduct an empirical geometric validation of our hypothesis in support of which we obtain overwhelming evidence. Subsequently, as our third contribution and based on our theory, we develop artificial mechanisms that leverage during learning the hypothesized (and verified) effect phonetics. We achieve significant and consistent improvements overall language pairs and datasets: French-English, German-English, and Chinese-English in medium task IWSLT'17 and French-English in large task WMT'18 Bio, with up to 4 BLEU points over the state-of-the-art. Moreover, our approaches are more robust than baselines when evaluated on unknown out-of-domain test sets with up to a 5 BLEU point increase.

arXiv:1911.04283

Data Efficient Direct Speech-to-Text Translation with Modality Agnostic Meta-Learning

End-to-end Speech Translation (ST) models have several advantages such as lower latency, smaller model size, and less error compounding over conventional pipelines that combine Automatic Speech Recognition (ASR) and text Machine Translation (MT) models. However, collecting large amounts of parallel data for ST task is more difficult compared to the ASR and MT tasks. Previous studies have proposed the use of transfer learning approaches to overcome the above difficulty. These approaches benefit from weakly supervised training data, such as ASR speech-to-transcript or MT text-to-text translation pairs. However, the parameters in these models are updated independently of each task, which may lead to sub-optimal solutions. In this work, we adopt a meta-learning algorithm to train a modality agnostic multi-task model that transfers knowledge from source tasks=ASR+MT to target task=ST where ST task severely lacks data. In the meta-learning phase, the parameters of the model are exposed to vast amounts of speech transcripts (e.g., English ASR) and text translations (e.g., English-German MT). During this phase, parameters are updated in such a way to understand speech, text representations, the relation between them, as well as act as a good initialization point for the target ST task. We evaluate the proposed meta-learning approach for ST tasks on English-German (En-De) and English-French (En-Fr) language pairs from the Multilingual Speech Translation Corpus (MuST-C). Our method outperforms the previous transfer learning approaches and sets new state-of-the-art results for En-De and En-Fr ST tasks by obtaining 9.18, and 11.76 BLEU point improvements, respectively.

arXiv:1911.03362

Domain, Translationese and Noise in Synthetic Data for Neural Machine Translation

The quality of neural machine translation can be improved by leveraging additional monolingual resources to create synthetic training data. Source-side monolingual data can be (forward-)translated into the target language for self-training; target-side monolingual data can be back-translated. It has been widely reported that back-translation delivers superior results, but could this be due to artefacts in the test sets? We perform a case study using French-English news translation task and separate test sets based on their original languages. We show that forward translation delivers superior gains in terms of BLEU on sentences that were originally in the source language, complementing previous studies which show large improvements with back-translation on sentences that were originally in the target language. To better understand when and why forward and back-translation are effective, we study the role of domains, translationese, and noise. While translationese effects are well known to influence MT evaluation, we also find evidence that news data from different languages shows subtle domain differences, which is another explanation for varying performance on different portions of the test set. We perform additional low-resource experiments which demonstrate that forward translation is more sensitive to the quality of the initial translation system than back-translation, and tends to perform worse in low-resource settings.

arXiv:1911.02971

Probing Contextualized Sentence Representations with Visual Awareness

We present a universal framework to model contextualized sentence representations with visual awareness that is motivated to overcome the shortcomings of the multimodal parallel data with manual annotations. For each sentence, we first retrieve a diversity of images from a shared cross-modal embedding space, which is pre-trained on a large-scale of text-image pairs. Then, the texts and images are respectively encoded by transformer encoder and convolutional neural network. The two sequences of representations are further fused by a simple and effective attention layer. The architecture can be easily applied to text-only natural language processing tasks without manually annotating multimodal parallel corpora. We apply the proposed method on three tasks, including neural machine translation, natural language inference and sequence labeling and experimental results verify the effectiveness.

arXiv:1911.02898

The LIG system for the English-Czech Text Translation Task of IWSLT 2019

In this paper, we present our submission for the English to Czech Text Translation Task of IWSLT 2019. Our system aims to study how pre-trained language models, used as input embeddings, can improve a specialized machine translation system trained on few data. Therefore, we implemented a Transformer-based encoder-decoder neural system which is able to use the output of a pre-trained language model as input embeddings, and we compared its performance under three configurations: 1) without any pre-trained language model (constrained), 2) using a language model trained on the monolingual parts of the allowed English-Czech data (constrained), and 3) using a language model trained on a large quantity of external monolingual data (unconstrained). We used BERT as external pre-trained language model (configuration 3), and BERT architecture for training our own language model (configuration 2). Regarding the training data, we trained our MT system on a small quantity of parallel text: one set only consists of the provided MuST-C corpus, and the other set consists of the MuST-C corpus and the News Commentary corpus from WMT. We observed that using the external pre-trained BERT improves the scores of our system by +0.8 to +1.5 of BLEU on our development set, and +0.97 to +1.94 of BLEU on the test set. However, using our own language model trained only on the allowed parallel data seems to improve the machine translation performances only when the system is trained on the smallest dataset.

arXiv:1911.02825

Improving Grammatical Error Correction with Machine Translation Pairs

We propose a novel data synthesis method to generate diverse error-corrected sentence pairs for improving grammatical error correction, which is based on a pair of machine translation models of different qualities (i.e., poor and good). The poor translation model resembles the ESL (English as a second language) learner and tends to generate translations of low quality in terms of fluency and grammatical correctness, while the good translation model generally generates fluent and grammatically correct translations. We build the poor and good translation model with phrase-based statistical machine translation model with decreased language model weight and neural machine translation model respectively. By taking the pair of their translations of the same sentences in a bridge language as error-corrected sentence pairs, we can construct unlimited pseudo parallel data. Our approach is capable of generating diverse fluency-improving patterns without being limited by the pre-defined rule set and the seed error-corrected data. Experimental results demonstrate the effectiveness of our approach and show that it can be combined with other synthetic data sources to yield further improvements.

arXiv:1911.02808공식 코드 · Papers with Code

Transition-Based Deep Input Linearization

Traditional methods for deep NLG adopt pipeline approaches comprising stages such as constructing syntactic input, predicting function words, linearizing the syntactic input and generating the surface forms. Though easier to visualize, pipeline approaches suffer from error propagation. In addition, information available across modules cannot be leveraged by all modules. We construct a transition-based model to jointly perform linearization, function word prediction and morphological generation, which considerably improves upon the accuracy compared to a pipelined baseline system. On a standard deep input linearization shared task, our system achieves the best results reported so far.

arXiv:1911.02737

SubCharacter Chinese-English Neural Machine Translation with Wubi encoding

Neural machine translation (NMT) is one of the best methods for understanding the differences in semantic rules between two languages. Especially for Indo-European languages, subword-level models have achieved impressive results. However, when the translation task involves Chinese, semantic granularity remains at the word and character level, so there is still need more fine-grained translation model of Chinese. In this paper, we introduce a simple and effective method for Chinese translation at the sub-character level. Our approach uses the Wubi method to translate Chinese into English; byte-pair encoding (BPE) is then applied. Our method for Chinese-English translation eliminates the need for a complicated word segmentation algorithm during preprocessing. Furthermore, our method allows for sub-character-level neural translation based on recurrent neural network (RNN) architecture, without preprocessing. The empirical results show that for Chinese-English translation tasks, our sub-character-level model has a comparable BLEU score to the subword model, despite having a much smaller vocabulary. Additionally, the small vocabulary is highly advantageous for NMT model compression.

arXiv:1911.02727

Understanding Knowledge Distillation in Non-autoregressive Machine Translation

Non-autoregressive machine translation (NAT) systems predict a sequence of output tokens in parallel, achieving substantial improvements in generation speed compared to autoregressive models. Existing NAT models usually rely on the technique of knowledge distillation, which creates the training data from a pretrained autoregressive model for better performance. Knowledge distillation is empirically useful, leading to large gains in accuracy for NAT models, but the reason for this success has, as of yet, been unclear. In this paper, we first design systematic experiments to investigate why knowledge distillation is crucial to NAT training. We find that knowledge distillation can reduce the complexity of data sets and help NAT to model the variations in the output data. Furthermore, a strong correlation is observed between the capacity of an NAT model and the optimal complexity of the distilled data for the best translation quality. Based on these findings, we further propose several approaches that can alter the complexity of data sets to improve the performance of NAT models. We achieve the state-of-the-art performance for the NAT-based models, and close the gap with the autoregressive baseline on WMT14 En-De benchmark.

arXiv:1911.02709

Using Interlinear Glosses as Pivot in Low-Resource Multilingual Machine Translation

We demonstrate a new approach to Neural Machine Translation (NMT) for low-resource languages using a ubiquitous linguistic resource, Interlinear Glossed Text (IGT). IGT represents a non-English sentence as a sequence of English lemmas and morpheme labels. As such, it can serve as a pivot or interlingua for NMT. Our contribution is four-fold. Firstly, we pool IGT for 1,497 languages in ODIN (54,545 glosses) and 70,918 glosses in Arapaho and train a gloss-to-target NMT system from IGT to English, with a BLEU score of 25.94. We introduce a multilingual NMT model that tags all glossed text with gloss-source language tags and train a universal system with shared attention across 1,497 languages. Secondly, we use the IGT gloss-to-target translation as a key step in an English-Turkish MT system trained on only 865 lines from ODIN. Thirdly, we we present five metrics for evaluating extremely low-resource translation when BLEU is no longer sufficient and evaluate the Turkish low-resource system using BLEU and also using accuracy of matching nouns, verbs, agreement, tense, and spurious repetition, showing large improvements.

arXiv:1911.02692

Multi-Domain Neural Machine Translation with Word-Level Adaptive Layer-wise Domain Mixing

Many multi-domain neural machine translation (NMT) models achieve knowledge transfer by enforcing one encoder to learn shared embedding across domains. However, this design lacks adaptation to individual domains. To overcome this limitation, we propose a novel multi-domain NMT model using individual modules for each domain, on which we apply word-level, adaptive and layer-wise domain mixing. We first observe that words in a sentence are often related to multiple domains. Hence, we assume each word has a domain proportion, which indicates its domain preference. Then word representations are obtained by mixing their embedding in individual domains based on their domain proportions. We show this can be achieved by carefully designing multi-head dot-product attention modules for different domains, and eventually taking weighted averages of their parameters by word-level layer-wise domain proportions. Through this, we can achieve effective domain knowledge sharing, and capture fine-grained domain-specific knowledge as well. Our experiments show that our proposed model outperforms existing ones in several NMT tasks.

arXiv:1911.03167

Europarl-ST: A Multilingual Corpus For Speech Translation Of Parliamentary Debates

Current research into spoken language translation (SLT),or speech-to-text translation, is often hampered by the lack of specific data resources for this task, as currently available SLT datasets are restricted to a limited set of language pairs. In this paper we present Europarl-ST, a novel multilingual SLT corpus containing paired audio-text samples for SLT from and into 6 European languages, for a total of 30 different translation directions. This corpus has been compiled using the debates held in the European Parliament in the period between 2008 and 2012. This paper describes the corpus creation process and presents a series of automatic speech recognition, machine translation and spoken language translation experiments that highlight the potential of this new resource. The corpus is released under a Creative Commons license and is freely accessible and downloadable.

arXiv:1911.03154

A General Framework for Adaptation of Neural Machine Translation to Simultaneous Translation

Despite the success of neural machine translation (NMT), simultaneous neural machine translation (SNMT), the task of translating in real time before a full sentence has been observed, remains challenging due to the syntactic structure difference and simultaneity requirements. In this paper, we propose a general framework for adapting neural machine translation to translate simultaneously. Our framework contains two parts: prefix translation that utilizes a consecutive NMT model to translate source prefixes and a stopping criterion that determines when to stop the prefix translation. Experiments on three translation corpora and two language pairs show the efficacy of the proposed framework on balancing the quality and latency in adapting NMT to perform simultaneous translation.

arXiv:1911.03110

Pretrained Language Models for Document-Level Neural Machine Translation

Previous work on document-level NMT usually focuses on limited contexts because of degraded performance on larger contexts. In this paper, we investigate on using large contexts with three main contributions: (1) Different from previous work which pertrained models on large-scale sentence-level parallel corpora, we use pretrained language models, specifically BERT, which are trained on monolingual documents; (2) We propose context manipulation methods to control the influence of large contexts, which lead to comparable results on systems using small and large contexts; (3) We introduce a multi-task training for regularization to avoid models overfitting our training corpora, which further improves our systems together with a deeper encoder. Experiments are conducted on the widely used IWSLT data sets with three language pairs, i.e., Chinese--English, French--English and Spanish--English. Results show that our systems are significantly better than three previously reported document-level systems.

arXiv:1911.03109공식 코드 · Papers with Code

Domain Robustness in Neural Machine Translation

Translating text that diverges from the training domain is a key challenge for machine translation. Domain robustness---the generalization of models to unseen test domains---is low for both statistical (SMT) and neural machine translation (NMT). In this paper, we study the performance of SMT and NMT models on out-of-domain test sets. We find that in unknown domains, SMT and NMT suffer from very different problems: SMT systems are mostly adequate but not fluent, while NMT systems are mostly fluent, but not adequate. For NMT, we identify such hallucinations (translations that are fluent but unrelated to the source) as a key reason for low domain robustness. To mitigate this problem, we empirically compare methods that are reported to improve adequacy or in-domain robustness in terms of their effectiveness at improving domain robustness. In experiments on German to English OPUS data, and German to Romansh (a low-resource setting) we find that several methods improve domain robustness. While those methods do lead to higher BLEU scores overall, they only slightly increase the adequacy of translations compared to SMT.

arXiv:1911.02215

Guiding Non-Autoregressive Neural Machine Translation Decoding with Reordering Information

Non-autoregressive neural machine translation (NAT) generates each target word in parallel and has achieved promising inference acceleration. However, existing NAT models still have a big gap in translation quality compared to autoregressive neural machine translation models due to the enormous decoding space. To address this problem, we propose a novel NAT framework named ReorderNAT which explicitly models the reordering information in the decoding procedure. We further introduce deterministic and non-deterministic decoding strategies that utilize reordering information to narrow the decoding search space in our proposed ReorderNAT. Experimental results on various widely-used datasets show that our proposed model achieves better performance compared to existing NAT models, and even achieves comparable translation quality as autoregressive translation models with a significant speedup.

arXiv:1911.01600

Integrating Dictionary Feature into A Deep Learning Model for Disease Named Entity Recognition

In recent years, Deep Learning (DL) models are becoming important due to their demonstrated success at overcoming complex learning problems. DL models have been applied effectively for different Natural Language Processing (NLP) tasks such as part-of-Speech (PoS) tagging and Machine Translation (MT). Disease Named Entity Recognition (Disease-NER) is a crucial task which aims at extracting disease Named Entities (NEs) from text. In this paper, a DL model for Disease-NER using dictionary information is proposed and evaluated on National Center for Biotechnology Information (NCBI) disease corpus and BC5CDR dataset. Word embeddings trained over general domain texts as well as biomedical texts have been used to represent input to the proposed model. This study also compares two different Segment Representation (SR) schemes, namely IOB2 and IOBES for Disease-NER. The results illustrate that using dictionary information, pre-trained word embeddings, character embeddings and CRF with global score improves the performance of Disease-NER system.

arXiv:1911.01597

Improving Bidirectional Decoding with Dynamic Target Semantics in Neural Machine Translation

Generally, Neural Machine Translation models generate target words in a left-to-right (L2R) manner and fail to exploit any future (right) semantics information, which usually produces an unbalanced translation. Recent works attempt to utilize the right-to-left (R2L) decoder in bidirectional decoding to alleviate this problem. In this paper, we propose a novel \textbf{D}ynamic \textbf{I}nteraction \textbf{M}odule (\textbf{DIM}) to dynamically exploit target semantics from R2L translation for enhancing the L2R translation quality. Different from other bidirectional decoding approaches, DIM firstly extracts helpful target information through addressing and reading operations, then updates target semantics for tracking the interactive history. Additionally, we further introduce an \textbf{agreement regularization} term into the training objective to narrow the gap between L2R and R2L translations. Experimental results on NIST Chinese$\Rightarrow$English and WMT'16 English$\Rightarrow$Romanian translation tasks show that our system achieves significant improvements over baseline systems, which also reaches comparable results compared to the state-of-the-art Transformer model with much fewer parameters of it.