PyTorchKR Papers

최신 논문

총 1,781편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2102.11387공식 코드 · Papers with Code

Exploiting Multimodal Reinforcement Learning for Simultaneous Machine Translation

This paper addresses the problem of simultaneous machine translation (SiMT) by exploring two main concepts: (a) adaptive policies to learn a good trade-off between high translation quality and low latency; and (b) visual information to support this process by providing additional (visual) contextual information which may be available before the textual input is produced. For that, we propose a multimodal approach to simultaneous machine translation using reinforcement learning, with strategies to integrate visual and textual information in both the agent and the environment. We provide an exploration on how different types of visual information and integration strategies affect the quality and latency of simultaneous translation models, and demonstrate that visual cues lead to higher quality while keeping the latency low.

arXiv:2102.10243공식 코드 · Papers with Code

Machine Translation Customization via Automatic Training Data Selection from the Web

Machine translation (MT) systems, especially when designed for an industrial setting, are trained with general parallel data derived from the Web. Thus, their style is typically driven by word/structure distribution coming from the average of many domains. In contrast, MT customers want translations to be specialized to their domain, for which they are typically able to provide text samples. We describe an approach for customizing MT systems on specific domains by selecting data similar to the target customer data to train neural translation models. We build document classifiers using monolingual target data, e.g., provided by the customers to select parallel training data from Web crawled data. Finally, we train MT models on our automatically selected data, obtaining a system specialized to the target domain. We tested our approach on the benchmark from WMT-18 Translation Task for News domains enabling comparisons with state-of-the-art MT systems. The results show that our models outperform the top systems while using less data and smaller models.

arXiv:2102.10160공식 코드 · Papers with Code

Multi-Domain Adaptation in Neural Machine Translation Through Multidimensional Tagging

While NMT has achieved remarkable results in the last 5 years, production systems come with strict quality requirements in arbitrarily niche domains that are not always adequately covered by readily available parallel corpora. This is typically addressed by training domain specific models, using fine-tuning methods and some variation of back-translation on top of in-domain monolingual corpora. However, industrial practitioners can rarely afford to focus on a single domain. A far more typical scenario includes a set of closely related, yet succinctly different sub-domains. At Booking.com, we need to translate property descriptions, user reviews, as well as messages, (for example those sent between a customer and an agent or property manager). An editor might need to translate articles across a set of different topics. An e-commerce platform would typically need to translate both the description of each item and the user generated content related to them. To this end, we propose MDT: a novel method to simultaneously fine-tune on several sub-domains by passing multidimensional sentence-level information to the model during training and inference. We show that MDT achieves results competitive to N specialist models each fine-tuned on a single constituent domain, while effectively serving all N sub-domains, therefore cutting development and maintenance costs by the same factor. Besides BLEU (industry standard automatic evaluation metric known to only weakly correlate with human judgement) we also report rigorous human evaluation results for all models and sub-domains as well as specific examples that better contextualise the performance of each model in terms of adequacy and fluency. To facilitate further research, we plan to make the code available upon acceptance.

arXiv:2102.11174공식 코드 · Papers with Code

Linear Transformers Are Secretly Fast Weight Programmers

We show the formal equivalence of linearised self-attention mechanisms and fast weight controllers from the early '90s, where a ``slow" neural net learns by gradient descent to program the ``fast weights" of another net through sequences of elementary programming instructions which are additive outer products of self-invented activation patterns (today called keys and values). Such Fast Weight Programmers (FWPs) learn to manipulate the contents of a finite memory and dynamically interact with it. We infer a memory capacity limitation of recent linearised softmax attention variants, and replace the purely additive outer products by a delta rule-like programming instruction, such that the FWP can more easily learn to correct the current mapping from keys to values. The FWP also learns to compute dynamically changing learning rates. We also propose a new kernel function to linearise attention which balances simplicity and effectiveness. We conduct experiments on synthetic retrieval problems as well as standard machine translation and language modelling tasks which demonstrate the benefits of our methods.

arXiv:2102.09377공식 코드 · Papers with Code

Learning Skill Equivalencies Across Platform Taxonomies

Assessment and reporting of skills is a central feature of many digital learning platforms. With students often using multiple platforms, cross-platform assessment has emerged as a new challenge. While technologies such as Learning Tools Interoperability (LTI) have enabled communication between platforms, reconciling the different skill taxonomies they employ has not been solved at scale. In this paper, we introduce and evaluate a methodology for finding and linking equivalent skills between platforms by utilizing problem content as well as the platform's clickstream data. We propose six models to represent skills as continuous real-valued vectors and leverage machine translation to map between skill spaces. The methods are tested on three digital learning platforms: ASSISTments, Khan Academy, and Cognitive Tutor. Our results demonstrate reasonable accuracy in skill equivalency prediction from a fine-grained taxonomy to a coarse-grained one, achieving an average recall@5 of 0.8 between the three platforms. Our skill translation approach has implications for aiding in the tedious, manual process of taxonomy to taxonomy mapping work, also called crosswalks, within the tutoring as well as standardized testing worlds.

arXiv:2102.08934공식 코드 · Papers with Code

Sparsely Factored Neural Machine Translation

The standard approach to incorporate linguistic information to neural machine translation systems consists in maintaining separate vocabularies for each of the annotated features to be incorporated (e.g. POS tags, dependency relation label), embed them, and then aggregate them with each subword in the word they belong to. This approach, however, cannot easily accommodate annotation schemes that are not dense for every word. We propose a method suited for such a case, showing large improvements in out-of-domain data, and comparable quality for the in-domain data. Experiments are performed in morphologically-rich languages like Basque and German, for the case of low-resource scenarios.

arXiv:2102.08597공식 코드 · Papers with Code

Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters

Recent works have demonstrated reasonable success of representation learning in hypercomplex space. Specifically, "fully-connected layers with Quaternions" (4D hypercomplex numbers), which replace real-valued matrix multiplications in fully-connected layers with Hamilton products of Quaternions, both enjoy parameter savings with only 1/4 learnable parameters and achieve comparable performance in various applications. However, one key caveat is that hypercomplex space only exists at very few predefined dimensions (4D, 8D, and 16D). This restricts the flexibility of models that leverage hypercomplex multiplications. To this end, we propose parameterizing hypercomplex multiplications, allowing models to learn multiplication rules from data regardless of whether such rules are predefined. As a result, our method not only subsumes the Hamilton product, but also learns to operate on any arbitrary nD hypercomplex space, providing more architectural flexibility using arbitrarily $1/n$ learnable parameters compared with the fully-connected layer counterpart. Experiments of applications to the LSTM and Transformer models on natural language inference, machine translation, text style transfer, and subject verb agreement demonstrate architectural flexibility and effectiveness of the proposed approach.

arXiv:2102.07847공식 코드 · Papers with Code

Meta Back-translation

Back-translation is an effective strategy to improve the performance of Neural Machine Translation~(NMT) by generating pseudo-parallel data. However, several recent works have found that better translation quality of the pseudo-parallel data does not necessarily lead to better final translation models, while lower-quality but more diverse data often yields stronger results. In this paper, we propose a novel method to generate pseudo-parallel data from a pre-trained back-translation model. Our method is a meta-learning algorithm which adapts a pre-trained back-translation model so that the pseudo-parallel data it generates would train a forward-translation model to do well on a validation set. In our evaluations in both the standard datasets WMT En-De'14 and WMT En-Fr'14, as well as a multilingual translation setting, our method leads to significant improvements over strong baselines. Our code will be made available.

arXiv:2102.08098공식 코드 · Papers with Code

GradInit: Learning to Initialize Neural Networks for Stable and Efficient Training

Innovations in neural architectures have fostered significant breakthroughs in language modeling and computer vision. Unfortunately, novel architectures often result in challenging hyper-parameter choices and training instability if the network parameters are not properly initialized. A number of architecture-specific initialization schemes have been proposed, but these schemes are not always portable to new architectures. This paper presents GradInit, an automated and architecture agnostic method for initializing neural networks. GradInit is based on a simple heuristic; the norm of each network layer is adjusted so that a single step of SGD or Adam with prescribed hyperparameters results in the smallest possible loss value. This adjustment is done by introducing a scalar multiplier variable in front of each parameter block, and then optimizing these variables using a simple numerical scheme. GradInit accelerates the convergence and test performance of many convolutional architectures, both with or without skip connections, and even without normalization layers. It also improves the stability of the original Transformer architecture for machine translation, enabling training it without learning rate warmup using either Adam or SGD under a wide range of learning rates and momentum coefficients. Code is available at https://github.com/zhuchen03/gradinit.

arXiv:2102.08220공식 코드 · Papers with Code

Non-Autoregressive Text Generation with Pre-trained Language Models

Non-autoregressive generation (NAG) has recently attracted great attention due to its fast inference speed. However, the generation quality of existing NAG models still lags behind their autoregressive counterparts. In this work, we show that BERT can be employed as the backbone of a NAG model to greatly improve performance. Additionally, we devise mechanisms to alleviate the two common problems of vanilla NAG models: the inflexibility of prefixed output length and the conditional independence of individual token predictions. Lastly, to further increase the speed advantage of the proposed model, we propose a new decoding strategy, ratio-first, for applications where the output lengths can be approximately estimated beforehand. For a comprehensive evaluation, we test the proposed model on three text generation tasks, including text summarization, sentence compression and machine translation. Experimental results show that our model significantly outperforms existing non-autoregressive baselines and achieves competitive performance with many strong autoregressive models. In addition, we also conduct extensive analysis experiments to reveal the effect of each proposed component.

arXiv:2102.06991공식 코드 · Papers with Code

The first large scale collection of diverse Hausa language datasets

Hausa language belongs to the Afroasiatic phylum, and with more first-language speakers than any other sub-Saharan African language. With a majority of its speakers residing in the Northern and Southern areas of Nigeria and the Republic of Niger, respectively, it is estimated that over 100 million people speak the language. Hence, making it one of the most spoken Chadic language. While Hausa is considered well-studied and documented language among the sub-Saharan African languages, it is viewed as a low resource language from the perspective of natural language processing (NLP) due to limited resources to utilise in NLP-related tasks. This is common to most languages in Africa; thus, it is crucial to enrich such languages with resources that will support and speed the pace of conducting various downstream tasks to meet the demand of the modern society. While there exist useful datasets, notably from news sites and religious texts, more diversity is needed in the corpus. We provide an expansive collection of curated datasets consisting of both formal and informal forms of the language from refutable websites and online social media networks, respectively. The collection is large and more diverse than the existing corpora by providing the first and largest set of Hausa social media data posts to capture the peculiarities in the language. The collection also consists of a parallel dataset, which can be used for tasks such as machine translation with applications in areas such as the detection of spurious or inciteful online content. We describe the curation process -- from the collection, preprocessing and how to obtain the data -- and proffer some research problems that could be addressed using the data.

arXiv:2102.06320공식 코드 · Papers with Code

On Automatic Parsing of Log Records

Software log analysis helps to maintain the health of software solutions and ensure compliance and security. Existing software systems consist of heterogeneous components emitting logs in various formats. A typical solution is to unify the logs using manually built parsers, which is laborious. Instead, we explore the possibility of automating the parsing task by employing machine translation (MT). We create a tool that generates synthetic Apache log records which we used to train recurrent-neural-network-based MT models. Models' evaluation on real-world logs shows that the models can learn Apache log format and parse individual log records. The median relative edit distance between an actual real-world log record and the MT prediction is less than or equal to 28%. Thus, we show that log parsing using an MT approach is promising.

arXiv:2102.01417공식 코드 · Papers with Code

Two Demonstrations of the Machine Translation Applications to Historical Documents

We present our demonstration of two machine translation applications to historical documents. The first task consists in generating a new version of a historical document, written in the modern version of its original language. The second application is limited to a document's orthography. It adapts the document's spelling to modern standards in order to achieve an orthography consistency and accounting for the lack of spelling conventions. We followed an interactive, adaptive framework that allows the user to introduce corrections to the system's hypothesis. The system reacts to these corrections by generating a new hypothesis that takes them into account. Once the user is satisfied with the system's hypothesis and validates it, the system adapts its model following an online learning strategy. This system is implemented following a client-server architecture. We developed a website which communicates with the neural models. All code is open-source and publicly available. The demonstration is hosted at http://demosmt.prhlt.upv.es/mthd/.

arXiv:2102.00881공식 코드 · Papers with Code

Gamified Crowdsourcing for Idiom Corpora Construction

Learning idiomatic expressions is seen as one of the most challenging stages in second language learning because of their unpredictable meaning. A similar situation holds for their identification within natural language processing applications such as machine translation and parsing. The lack of high-quality usage samples exacerbates this challenge not only for humans but also for artificial intelligence systems. This article introduces a gamified crowdsourcing approach for collecting language learning materials for idiomatic expressions; a messaging bot is designed as an asynchronous multiplayer game for native speakers who compete with each other while providing idiomatic and nonidiomatic usage examples and rating other players' entries. As opposed to classical crowdprocessing annotation efforts in the field, for the first time in the literature, a crowdcreating & crowdrating approach is implemented and tested for idiom corpora construction. The approach is language independent and evaluated on two languages in comparison to traditional data preparation techniques in the field. The reaction of the crowd is monitored under different motivational means (namely, gamification affordances and monetary rewards). The results reveal that the proposed approach is powerful in collecting the targeted materials, and although being an explicit crowdsourcing approach, it is found entertaining and useful by the crowd. The approach has been shown to have the potential to speed up the construction of idiom corpora for different natural languages to be used as second language learning material, training data for supervised idiom identification systems, or samples for lexicographic studies.

arXiv:2101.12640공식 코드 · Papers with Code

Enhancing the Transformer Decoder with Transition-based Syntax

Notwithstanding recent advances, syntactic generalization remains a challenge for text decoders. While some studies showed gains from incorporating source-side symbolic syntactic and semantic structure into text generation Transformers, very little work addressed the decoding of such structure. We propose a general approach for tree decoding using a transition-based approach. Examining the challenging test case of incorporating Universal Dependencies syntax into machine translation, we present substantial improvements on test sets that focus on syntactic generalization, while presenting improved or comparable performance on standard MT benchmarks. Further qualitative analysis addresses cases where syntactic generalization in the vanilla Transformer decoder is inadequate and demonstrates the advantages afforded by integrating syntactic information.

arXiv:2101.10035공식 코드 · Papers with Code

Facilitating Terminology Translation with Target Lemma Annotations

Most of the recent work on terminology integration in machine translation has assumed that terminology translations are given already inflected in forms that are suitable for the target language sentence. In day-to-day work of professional translators, however, it is seldom the case as translators work with bilingual glossaries where terms are given in their dictionary forms; finding the right target language form is part of the translation process. We argue that the requirement for apriori specified target language forms is unrealistic and impedes the practical applicability of previous work. In this work, we propose to train machine translation systems using a source-side data augmentation method that annotates randomly selected source language words with their target language lemmas. We show that systems trained on such augmented data are readily usable for terminology integration in real-life translation scenarios. Our experiments on terminology translation into the morphologically complex Baltic and Uralic languages show an improvement of up to 7 BLEU points over baseline systems with no means for terminology integration and an average improvement of 4 BLEU points over the previous work. Results of the human evaluation indicate a 47.7% absolute improvement over the previous work in term translation accuracy when translating into Latvian.

arXiv:2101.06561공식 코드 · Papers with Code

GENIE: Toward Reproducible and Standardized Human Evaluation for Text Generation

While often assumed a gold standard, effective human evaluation of text generation remains an important, open area for research. We revisit this problem with a focus on producing consistent evaluations that are reproducible -- over time and across different populations. We study this goal in different stages of the human evaluation pipeline. In particular, we consider design choices for the annotation interface used to elicit human judgments and their impact on reproducibility. Furthermore, we develop an automated mechanism for maintaining annotator quality via a probabilistic model that detects and excludes noisy annotators. Putting these lessons together, we introduce GENIE: a system for running standardized human evaluations across different generation tasks. We instantiate GENIE with datasets representing four core challenges in text generation: machine translation, summarization, commonsense reasoning, and machine comprehension. For each task, GENIE offers a leaderboard that automatically crowdsources annotations for submissions, evaluating them along axes such as correctness, conciseness, and fluency. We have made the GENIE leaderboards publicly available, and have already ranked 50 submissions from 10 different research groups. We hope GENIE encourages further progress toward effective, standardized evaluations for text generation.

arXiv:2101.06125공식 코드 · Papers with Code

The Impact of Post-editing and Machine Translation on Creativity and Reading Experience

This article presents the results of a study involving the translation of a fictional story from English into Catalan in three modalities: machine-translated (MT), post-edited (MTPE) and translated without aid (HT). Each translation was analysed to evaluate its creativity. Subsequently, a cohort of 88 Catalan participants read the story in a randomly assigned modality and completed a survey. The results show that HT presented a higher creativity score if compared to MTPE and MT. HT also ranked higher in narrative engagement, and translation reception, while MTPE ranked marginally higher in enjoyment. HT and MTPE show no statistically significant differences in any category, whereas MT does in all variables tested. We conclude that creativity is highest when professional translators intervene in the process, especially when working without any aid. We hypothesize that creativity in translation could be the factor that enhances reading engagement and the reception of translated literary texts.

arXiv:2101.01761공식 코드 · Papers with Code

AutoDropout: Learning Dropout Patterns to Regularize Deep Networks

Neural networks are often over-parameterized and hence benefit from aggressive regularization. Conventional regularization methods, such as Dropout or weight decay, do not leverage the structures of the network's inputs and hidden states. As a result, these conventional methods are less effective than methods that leverage the structures, such as SpatialDropout and DropBlock, which randomly drop the values at certain contiguous areas in the hidden states and setting them to zero. Although the locations of dropout areas random, the patterns of SpatialDropout and DropBlock are manually designed and fixed. Here we propose to learn the dropout patterns. In our method, a controller learns to generate a dropout pattern at every channel and layer of a target network, such as a ConvNet or a Transformer. The target network is then trained with the dropout pattern, and its resulting validation performance is used as a signal for the controller to learn from. We show that this method works well for both image recognition on CIFAR-10 and ImageNet, as well as language modeling on Penn Treebank and WikiText-2. The learned dropout patterns also transfers to different tasks and datasets, such as from language model on Penn Treebank to Engligh-French translation on WMT 2014. Our code will be available.

arXiv:2101.00421공식 코드 · Papers with Code

The Highs and Lows of Simple Lexical Domain Adaptation Approaches for Neural Machine Translation

Machine translation systems are vulnerable to domain mismatch, especially in a low-resource scenario. Out-of-domain translations are often of poor quality and prone to hallucinations, due to exposure bias and the decoder acting as a language model. We adopt two approaches to alleviate this problem: lexical shortlisting restricted by IBM statistical alignments, and hypothesis re-ranking based on similarity. The methods are computationally cheap, widely known, but not extensively experimented on domain adaptation. We demonstrate success on low-resource out-of-domain test sets, however, the methods are ineffective when there is sufficient data or too great domain mismatch. This is due to both the IBM model losing its advantage over the implicitly learned neural alignment, and issues with subword segmentation of out-of-domain words.