Paper Digest: Recent Papers on Machine Translation
Paper Digest Team extracted all recent Machine Translation related papers on our radar, and generated highlight sentences for them. The results are then sorted by relevance & date. In addition to this ‘static’ page, we also provide a real-time version of this article, which has more coverage and is updated in real time to include the most recent updates on this topic.
If you do not want to miss any interesting academic paper, you are welcome to sign up our free daily paper digest service to get updates on new papers published in your area every day. You are also welcome to follow us on Twitter and Linkedin to get updated with new conference digests.
Paper Digest Team
team@paperdigest.org
TABLE 1: Paper Digest: Recent Papers on Machine Translation
| Paper | Author(s) | Source | Date | |
|---|---|---|---|---|
| 1 | UPC’s Speech Translation System for IWSLT 2021 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes the submission to the IWSLT 2021 offline speech translation task by the UPC Machine Translation group. |
Gerard I. Gállego; Ioannis Tsiamas; Carlos Escolano; José A. R. Fonollosa; Marta R. Costa-jussà; | arxiv-cs.CL | 2021-05-10 |
| 2 | Self-Guided Curriculum Learning for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Inspired by this, we propose a self-guided curriculum strategy to encourage the learning of neural machine translation (NMT) models to follow the above recovery criterion, where we cast the recovery degree of each training example as its learning difficulty. |
Lei Zhou; Liang Ding; Kevin Duh; Ryohei Sasano; Koichi Takeda; | arxiv-cs.CL | 2021-05-10 |
| 3 | Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a continual pre-training (CPT) framework on mBART to effectively adapt it to unseen languages. |
Zihan Liu; Genta Indra Winata; Pascale Fung; | arxiv-cs.CL | 2021-05-09 |
| 4 | Learning Shared Semantic Space for Speech-to-Text Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In observation of this obstacle, we propose to bridge this representation gap with Chimera. |
Chi Han; Mingxuan Wang; Heng Ji; Lei Li; | arxiv-cs.CL | 2021-05-07 |
| 5 | Measuring and Increasing Context Usage in Context-Aware Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we introduce a new metric, conditional cross-mutual information, to quantify the usage of context by these models. |
Patrick Fernandes; Kayo Yin; Graham Neubig; André F. T. Martins; | arxiv-cs.CL | 2021-05-07 |
| 6 | Duplex Sequence-to-Sequence Learning for Reversible Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address this issue, we propose a {\em duplex} seq2seq neural network, REDER (Reversible Duplex Transformer), and apply it to machine translation. |
ZAIXIANG ZHENG et. al. | arxiv-cs.CL | 2021-05-07 |
| 7 | Aligning Subtitles in Sign Language Videos Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: The goal of this work is to temporally align asynchronous subtitles in sign language videos. |
HANNAH BULL et. al. | arxiv-cs.CV | 2021-05-06 |
| 8 | Data Augmentation By Concatenation for Low-Resource Translation: A Mystery and A Solution Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we investigate the driving factors behind concatenation, a simple but effective data augmentation method for low-resource neural machine translation. |
Toan Q. Nguyen; Kenton Murray; David Chiang; | arxiv-cs.CL | 2021-05-04 |
| 9 | Searchable Hidden Intermediates for End-to-End Models of Decomposable Sequence Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present an end-to-end framework that exploits compositionality to learn searchable hidden representations at intermediate stages of a sequence model using decomposed sub-tasks. |
Siddharth Dalmia; Brian Yan; Vikas Raunak; Florian Metze; Shinji Watanabe; | arxiv-cs.CL | 2021-05-02 |
| 10 | Hidden Backdoors in Human-Centric Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we create covert and natural triggers for textual backdoor attacks, \textit{hidden backdoors}, where triggers can fool both modern language models and human inspection. |
SHAOFENG LI et. al. | arxiv-cs.CL | 2021-05-01 |
| 11 | Semantic Relation Preserving Knowledge Distillation for Image-to-Image Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a novel method to address this problem by applying knowledge distillation together with distillation of a semantic relation preserving matrix. |
Zeqi Li; Ruowei Jiang; Parham Aarabi; | arxiv-cs.CV | 2021-04-30 |
| 12 | Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: As a step toward this goal, we propose an evaluation methodology grounded in explicit error analysis, based on the Multidimensional Quality Metrics (MQM) framework. |
MARKUS FREITAG et. al. | arxiv-cs.CL | 2021-04-29 |
| 13 | Impact of Encoding and Segmentation Strategies on End-to-End Simultaneous Speech Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper investigates two key aspects of end-to-end simultaneous speech translation: (a) how to encode efficiently the continuous speech flow, and (b) how to segment the speech flow in order to alternate optimally between reading (R: encoding input) and writing (W: decoding output) operations. |
Ha Nguyen; Yannick Estève; Laurent Besacier; | arxiv-cs.CL | 2021-04-29 |
| 14 | Shellcode_IA32: A Dataset for Automatic Shellcode Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We take the first step to address the task of automatically generating shellcodes, i.e., small pieces of code used as a payload in the exploitation of a software vulnerability, starting from natural language comments. We assemble and release a novel dataset (Shellcode_IA32), consisting of challenging but common assembly instructions with their natural language descriptions. |
PIETRO LIGUORI et. al. | arxiv-cs.SE | 2021-04-27 |
| 15 | Good Artists Copy, Great Artists Steal: Model Extraction Attacks Against Image Translation Generative Adversarial Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a framework for conducting model extraction attacks against image translation models, and show that the adversary can successfully extract functional surrogate models. |
Sebastian Szyller; Vasisht Duddu; Tommi Gröndahl; N. Asokan; | arxiv-cs.LG | 2021-04-26 |
| 16 | Reranking Machine Translation Hypotheses with Structured and Web-based Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we investigate the use of linguistically motivated and computationally efficient structured language models for reranking N-best hypotheses in a statistical machine translation system. |
Wen Wang; Andreas Stolcke; Jing Zheng; | arxiv-cs.CL | 2021-04-25 |
| 17 | Automatic Post-Editing for Translating Chinese Novels to Vietnamese Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present the first attempt to tackle the APE task for Vietnamese. Specifically, we construct the first large-scale dataset of 5M Vietnamese translated and corrected sentence pairs. |
Thanh Vu; Dai Quoc Nguyen; | arxiv-cs.CL | 2021-04-25 |
| 18 | Modeling Coverage for Non-Autoregressive Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we argue that these issues of NAT can be addressed through coverage modeling, which has been proved to be useful in autoregressive decoding. |
Yong Shan; Yang Feng; Chenze Shao; | arxiv-cs.CL | 2021-04-24 |
| 19 | Should We Stop Training More Monolingual Models, and Simply Use Machine Translation Instead? Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Most work in NLP makes the assumption that it is desirable to develop solutions in the native language in question. |
Tim Isbister; Fredrik Carlsson; Magnus Sahlgren; | arxiv-cs.CL | 2021-04-21 |
| 20 | End-to-end Speech Translation Via Cross-modal Progressive Training Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose Cross Speech-Text Network (XSTNet), an end-to-end model for speech-to-text translation. |
Rong Ye; Mingxuan Wang; Lei Li; | arxiv-cs.CL | 2021-04-21 |
| 21 | Grammatical Error Generation Based on Translated Fragments Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our method aims at simulating mistakes made by second language learners, and produces a wider range of non-native style language in comparison to state-of-the-art synthetic data creation methods. |
Eetu Sjöblom; Mathias Creutz; Teemu Vahtola; | arxiv-cs.CL | 2021-04-20 |
| 22 | Addressing The Vulnerability of NMT in Input Perturbations Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we improve the robustness of NMT models by reducing the effect of noisy words through a Context-Enhanced Reconstruction (CER) approach. |
Weiwen Xu; Ai Ti Aw; Yang Ding; Kui Wu; Shafiq Joty; | arxiv-cs.CL | 2021-04-20 |
| 23 | Can Latent Alignments Improve Autoregressive Machine Translation? Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Latent alignment objectives such as CTC and AXE significantly improve non-autoregressive machine translation models. |
Adi Haviv; Lior Vassertail; Omer Levy; | arxiv-cs.CL | 2021-04-19 |
| 24 | Stream-level Latency Evaluation for Simultaneous Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work proposes a stream-level adaptation of the current latency measures based on a re-segmentation approach applied to the output translation, that is successfully evaluated on streaming conditions for a reference IWSLT task. |
Javier Iranzo-Sánchez; Jorge Civera; Alfons Juan; | arxiv-cs.CL | 2021-04-18 |
| 25 | Zero-shot Cross-lingual Transfer of Neural Machine Translation with Multilingual Pretrained Encoders Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we focus on a zero-shot cross-lingual transfer task in NMT. |
GUANHUA CHEN et. al. | arxiv-cs.CL | 2021-04-18 |
| 26 | DCH-2: A Parallel Customer-Helpdesk Dialogue Corpus with Distributions of Annotators’ Labels Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a data set called DCH-2, which contains 4,390 real customer-helpdesk dialogues in Chinese and their English translations. |
Zhaohao Zeng; Tetsuya Sakai; | arxiv-cs.CL | 2021-04-18 |
| 27 | Embedding-Enhanced Giza++: Improving Alignment in Low- and High- Resource Scenarios Using Embedding Space Geometry Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce Embedding-Enhanced Giza++, and outperform Giza++ without any of the aforementioned factors. |
Kelly Marchisio; Conghao Xiong; Philipp Koehn; | arxiv-cs.CL | 2021-04-18 |
| 28 | On The Strengths of Cross-Attention in Pretrained Transformers for Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We study the power of cross-attention in the Transformer architecture within the context of machine translation. |
Mozhdeh Gheini; Xiang Ren; Jonathan May; | arxiv-cs.CL | 2021-04-18 |
| 29 | XLEnt: Mining A Large Cross-lingual Entity Dataset with Lexical-Semantic-Phonetic Word Alignment Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address this, we propose Lexical-Semantic-Phonetic Align (LSP-Align), a technique to automatically mine cross-lingual entity lexicon from the web. |
Ahmed El-Kishky; Adi Renduchintala; James Cross; Francisco Guzmán; Philipp Koehn; | arxiv-cs.CL | 2021-04-17 |
| 30 | Improving Neural Machine Translation with Compact Word Embedding Tables Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we analyze the impact and utility of such matrices in the context of neural machine translation (NMT). |
Krtin Kumar; Mehdi Rezagholizadeh; Yiu Sing Lau; Qun Liu; | arxiv-cs.CL | 2021-04-17 |
| 31 | Sentence Concatenation Approach to Data Augmentation for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, this study proposes a simple data augmentation method to handle long sentences. |
Seiichiro Kondo; Kengo Hotate; Masahiro Kaneko; Mamoru Komachi; | arxiv-cs.CL | 2021-04-17 |
| 32 | Sentence Alignment with Parallel Documents Helps Biomedical Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work presents a new unsupervised sentence alignment method and explores features in training biomedical neural machine translation (NMT) systems. |
Shengxuan Luo; Huaiyuan Ying; Sheng Yu; | arxiv-cs.CL | 2021-04-17 |
| 33 | Crossing The Conversational Chasm: A Primer on Multilingual Task-Oriented Dialogue Systems Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we provide an extensive overview of existing efforts in multilingual ToD and analyse the factors preventing the development of truly multilingual ToD systems. |
Evgeniia Razumovskaia; Goran Glavaš; Olga Majewska; Anna Korhonen; Ivan Vulić; | arxiv-cs.CL | 2021-04-17 |
| 34 | MT6: Multilingual Pretrained Text-to-Text Transformer with Translation Pairs Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we improve multilingual text-to-text transfer Transformer with translation pairs (mT6). |
ZEWEN CHI et. al. | arxiv-cs.CL | 2021-04-17 |
| 35 | Multilingual and Cross-Lingual Intent Detection from Spoken Data Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a systematic study on multilingual and cross-lingual intent detection from spoken data. |
DANIELA GERZ et. al. | arxiv-cs.CL | 2021-04-17 |
| 36 | Robust Open-Vocabulary Translation from Visual Text Representations Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Motivated by the robustness of human language processing, we propose the use of visual text representations, which dispense with a finite set of text embeddings in favor of continuous vocabularies created by processing visually rendered text. |
Elizabeth Salesky; David Etter; Matt Post; | arxiv-cs.CL | 2021-04-16 |
| 37 | IndoNLG: Benchmark and Resources for Evaluating Indonesian Natural Language Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce IndoNLG, the first such benchmark for the Indonesian language for natural language generation (NLG). We provide a vast and clean pre-training corpus of Indonesian, Sundanese, and Javanese datasets called Indo4B-Plus, which is used to train our pre-trained NLG model, IndoBART. |
SAMUEL CAHYAWIJAYA et. al. | arxiv-cs.CL | 2021-04-16 |
| 38 | Context-Adaptive Document-Level Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we introduce a data-adaptive method that enables the model to adopt the necessary and useful context. |
Linlin Zhang; | arxiv-cs.CL | 2021-04-16 |
| 39 | Serial or Parallel? Plug-able Adapter for Multilingual Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address the two issues, we propose PAM, a Transformer model augmented with defusion adaptation for multilingual machine translation. |
Yaoming Zhu; Jiangtao Feng; Chengqi Zhao; Mingxuan Wang; Lei Li; | arxiv-cs.CL | 2021-04-16 |
| 40 | Towards Variable-Length Textual Adversarial Attacks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose variable-length textual adversarial attacks~(VL-Attack) and integrate three atomic operations, namely \textit{insertion}, \textit{deletion} and \textit{replacement}, into a unified framework, by introducing and manipulating a special \textit{blank} token while attacking. |
JUNLIANG GUO et. al. | arxiv-cs.CL | 2021-04-16 |
| 41 | Wikily Neural Machine Translation Tailored to Cross-Lingual Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a simple but effective approach for leveraging Wikipedia for neural machine translation as well as cross-lingual tasks of image captioning and dependency parsing without using any direct supervision from external parallel data or supervised models in the target language. |
Mohammad Sadegh Rasooli; Chris Callison-Burch; Derry Tanti Wijaya; | arxiv-cs.CL | 2021-04-16 |
| 42 | Improving Gender Translation Accuracy with Filtered Self-Training Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a gender-filtered self-training technique to improve gender translation accuracy on unambiguously gendered inputs. |
Prafulla Kumar Choubey; Anna Currey; Prashant Mathur; Georgiana Dinu; | arxiv-cs.CL | 2021-04-15 |
| 43 | Segmenting Subtitles for Correcting ASR Segmentation Errors Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a model for correcting the acoustic segmentation of ASR models for low-resource languages to improve performance on downstream tasks. |
DAVID WAN et. al. | arxiv-cs.CL | 2021-04-15 |
| 44 | Zero-Shot Cross-lingual Semantic Parsing Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a multi-task encoder-decoder model to transfer parsing knowledge to additional languages using only English-Logical form paired data and unlabeled, monolingual utterances in each test language. |
Tom Sherborne; Mirella Lapata; | arxiv-cs.CL | 2021-04-15 |
| 45 | Sometimes We Want Translationese Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we draw attention to the fact that for some applications, faithfulness to the original (input) text is important to preserve, even if it means introducing unusual language patterns in the (output) translation. |
Prasanna Parthasarathi; Koustuv Sinha; Joelle Pineau; Adina Williams; | arxiv-cs.CL | 2021-04-15 |
| 46 | Simultaneous Multi-Pivot Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To solve this issue, we propose multi-pivot translation and apply it to a simultaneous translation setting involving pivot languages. |
Raj Dabre; Aizhan Imankulova; Masahiro Kaneko; Abhisek Chakrabarty; | arxiv-cs.CL | 2021-04-15 |
| 47 | Adaptive Sparse Transformer for Multilingual Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Based on these insights, we propose an adaptive and sparse architecture for multilingual modeling, and train the model to learn shared and language-specific parameters to improve the positive transfer and mitigate the interference. |
Hongyu Gong; Xian Li; Dmitriy Genzel; | arxiv-cs.CL | 2021-04-15 |
| 48 | Hierarchical Learning for Generation with Long Source Sequences Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We design and study a new Hierarchical Attention Transformer-based architecture (HAT) that outperforms standard Transformers on several sequence to sequence tasks. |
Tobias Rohde; Xiaoxia Wu; Yinhan Liu; | arxiv-cs.CL | 2021-04-15 |
| 49 | Reward Optimization for Neural Machine Translation with Learned Metrics Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we investigate whether it is beneficial to optimize NMT models with the state-of-the-art model-based metric, BLEURT. |
Raphael Shu; Kang Min Yoo; Jung-Woo Ha; | arxiv-cs.CL | 2021-04-15 |
| 50 | Investigating Failures of Automatic Translation in The Case of Unambiguous Gender Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We find that even when the surrounding context provides unambiguous evidence of the appropriate grammatical gender marking, no transformer based model we tested was able to accurately gender occupation nouns systematically. We release an evaluation scheme and dataset for measuring the ability of transformer based NMT models to translate gender morphology correctly in unambiguous contexts across syntactically diverse sentences. |
Adithya Renduchintala; Adina Williams; | arxiv-cs.CL | 2021-04-15 |
| 51 | First The Worst: Finding Better Gender Translations During Beam Search Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We find that a combination of these techniques allows large gains in WinoMT accuracy without requiring additional bilingual data or an additional NMT model. |
Danielle Saunders; Rosie Sallis; Bill Byrne; | arxiv-cs.CL | 2021-04-15 |
| 52 | Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: The development of deep learning techniques has allowed Neural Machine Translation (NMT) models to become extremely powerful, given sufficient training data and training time. … |
Danielle Saunders; | arxiv-cs.CL | 2021-04-14 |
| 53 | I Wish I Would Have Loved This One, But I Didn’t — A Multilingual Dataset for Counterfactual Detection in Product Reviews Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We consider the problem of counterfactual detection (CFD) in product reviews. For this purpose, we annotate a multilingual CFD dataset from Amazon product reviews covering counterfactual statements written in English, German, and Japanese languages. |
James O’Neill; Polina Rozenshtein; Ryuichi Kiryo; Motoko Kubota; Danushka Bollegala; | arxiv-cs.CL | 2021-04-14 |
| 54 | The Curious Case of Hallucinations in Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we study hallucinations in Neural Machine Translation (NMT), which lie at an extreme end on the spectrum of NMT pathologies. |
Vikas Raunak; Arul Menezes; Marcin Junczys-Dowmunt; | arxiv-cs.CL | 2021-04-14 |
| 55 | Source and Target Bidirectional Knowledge Distillation for End-to-end Speech Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we focus on sequence-level knowledge distillation (SeqKD) from external text-based NMT models. |
Hirofumi Inaguma; Tatsuya Kawahara; Shinji Watanabe; | arxiv-cs.CL | 2021-04-13 |
| 56 | Gender Bias in Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: Machine translation (MT) technology has facilitated our daily tasks by providing accessible shortcuts for gathering, elaborating and communicating information. However, it can … |
Beatrice Savoldi; Marco Gaido; Luisa Bentivogli; Matteo Negri; Marco Turchi; | arxiv-cs.CL | 2021-04-13 |
| 57 | Restoring and Mining The Records of The Joseon Dynasty Via Neural Language Modeling and Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In response, we present a multi-task learning approach to restore and translate historical documents based on a self-attention mechanism, specifically utilizing two Korean historical records, ones of the most voluminous historical records in the world. |
KYEONGPIL KANG et. al. | arxiv-cs.CL | 2021-04-13 |
| 58 | Backtranslation Feedback Improves User Confidence in MT, Not Quality Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we describe an experiment on outbound translation from English to Czech and Estonian. |
VILÉM ZOUHAR et. al. | arxiv-cs.CL | 2021-04-12 |
| 59 | FUDGE: Controlled Text Generation With Future Discriminators Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose Future Discriminators for Generation (FUDGE), a flexible and modular method for controlled text generation. |
Kevin Yang; Dan Klein; | arxiv-cs.CL | 2021-04-12 |
| 60 | Machine Translation Decoding Beyond Beam Search Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our aim is to establish whether beam search can be replaced by a more powerful metric-driven search technique. |
RÉMI LEBLOND et. al. | arxiv-cs.CL | 2021-04-12 |
| 61 | Macro-Average: Rare Types Are Important Too Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We explore the simple type-based classifier metric, MacroF1, and study its applicability to MT evaluation. |
Thamme Gowda; Weiqiu You; Constantine Lignos; Jonathan May; | arxiv-cs.CL | 2021-04-12 |
| 62 | Family of Origin and Family of Choice: Massively Parallel Lexiconized Iterative Pretraining for Severely Low Resource Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To translate named entities correctly, we build a massive lexicon table for 2,939 Bible named entities in 124 source languages, and include many that occur once and covers more than 66 severely low resource languages. |
Zhong Zhou; Alex Waibel; | arxiv-cs.CL | 2021-04-12 |
| 63 | Assessing Reference-Free Peer Evaluation for Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We analyze various potential weaknesses of the approach and find that it is surprisingly robust and likely to offer reasonable performance across a broad spectrum of domains and different system qualities. |
Sweta Agrawal; George Foster; Markus Freitag; Colin Cherry; | arxiv-cs.CL | 2021-04-11 |
| 64 | UniDrop: A Simple Yet Effective Technique to Improve Transformer Without Extra Cost Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, in this paper, we integrate different dropout techniques into the training of Transformer models. |
ZHEN WU et. al. | arxiv-cs.CL | 2021-04-11 |
| 65 | Sentiment-based Candidate Selection for NMT Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Grounded in the observation that UGC features highly idiomatic, sentiment-charged language, we propose a decoder-side approach that incorporates automatic sentiment scoring into the MT candidate selection process. |
Alex Jones; Derry Tanti Wijaya; | arxiv-cs.CL | 2021-04-10 |
| 66 | Chinese Character Decomposition for Neural MT with Multi-Word Expressions Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Chinese character decomposition has been used as a feature to enhance Machine Translation (MT) models, combining radicals into character and word level models. |
Lifeng Han; Gareth J. F. Jones; Alan F. Smeaton; Paolo Bolzoni; | arxiv-cs.CL | 2021-04-09 |
| 67 | Design and Implementation of English To Yoruba Verb Phrase Machine Translation System Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We aim to develop an English to Yoruba machine translation system which can translate English verb phrase text to its Yoruba equivalent.Words from both languages Source Language and Target Language were collected for the verb phrase group in the home domain.The lexical translation is done by assigning values of the matching word in the dictionary.The syntax of the two languages was realized using Context-Free Grammar,we validated the rewrite rules with finite state automata.The human evaluation method was used and expert fluency scored.The evaluation shows the system performed better than that of sampled Google translation with over 70 percent of the response matching that of the system’s output. |
Safiriyu Eludiora; Benjamin Ajibade; | arxiv-cs.CL | 2021-04-08 |
| 68 | Perfection Not Required? Human-AI Partnerships in Code Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We examine the extent to which software engineers would tolerate such imperfections and explore ways to aid the detection and correction of those errors. |
JUSTIN D. WEISZ et. al. | arxiv-cs.HC | 2021-04-08 |
| 69 | Extended Parallel Corpus for Amharic-English Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes the acquisition, preprocessing, segmentation, and alignment of an Amharic-English parallel corpus. |
Andargachew Mekonnen Gezmu; Andreas Nürnberger; Tesfaye Bayu Bati; | arxiv-cs.CL | 2021-04-08 |
| 70 | Interpreting Verbal Metaphors By Paraphrasing Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we interpret metaphors with BERT and WordNet hypernyms and synonyms in an unsupervised manner, showing that our method significantly outperforms the state-of-the-art baseline. |
Rui Mao; Chenghua Lin; Frank Guerin; | arxiv-cs.CL | 2021-04-07 |
| 71 | AI4D — African Language Program Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work details the AI4D – African Language Program, a 3-part project that 1) incentivised the crowd-sourcing, collection and curation of language datasets through an online quantitative and qualitative challenge, 2) supported research fellows for a period of 3-4 months to create datasets annotated for NLP tasks, and 3) hosted competitive Machine Learning challenges on the basis of these datasets. |
KATHLEEN SIMINYU et. al. | arxiv-cs.CL | 2021-04-06 |
| 72 | Better Neural Machine Translation By Extracting Linguistic Information from BERT Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: Adding linguistic information (syntax or semantics) to neural machine translation (NMT) has mostly focused on using point estimates from pre-trained models. Directly using the … |
Hassan S. Shavarani; Anoop Sarkar; | arxiv-cs.CL | 2021-04-06 |
| 73 | ODE Transformer: An Ordinary Differential Equation-Inspired Model for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore a deeper relationship between Transformer and numerical methods of ODEs. |
BEI LI et. al. | arxiv-cs.CL | 2021-04-06 |
| 74 | IndT5: A Text-to-Text Transformer for 10 Indigenous Languages Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we introduce IndT5, the first Transformer language model for Indigenous languages. |
El Moatez Billah Nagoudi; Wei-Rui Chen; Muhammad Abdul-Mageed; Hasan Cavusogl; | arxiv-cs.CL | 2021-04-04 |
| 75 | Attention Forcing for Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Several approaches have been proposed to address this problem, normally by selectively using the generated output history. |
Qingyun Dou; Yiting Lu; Potsawee Manakul; Xixin Wu; Mark J. F. Gales; | arxiv-cs.CL | 2021-04-02 |
| 76 | Many-to-English Machine Translation Tools, Data, and Pretrained Models Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present useful tools for machine translation research: MTData, NLCodec, and RTG. |
Thamme Gowda; Zhao Zhang; Chris A Mattmann; Jonathan May; | arxiv-cs.CL | 2021-04-01 |
| 77 | The Spatially-Correlative Loss for Various Image Translation Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a novel spatially-correlative loss that is simple, efficient and yet effective for preserving scene structure consistency while supporting large appearance changes during unpaired image-to-image (I2I) translation. |
Chuanxia Zheng; Tat-Jen Cham; Jianfei Cai; | arxiv-cs.CV | 2021-04-01 |
| 78 | UC2: Universal Cross-lingual Cross-modal Vision-and-Language Pre-training Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To generalize this success to non-English languages, we introduce UC2, the first machine translation-augmented framework for cross-lingual cross-modal representation learning. |
MINGYANG ZHOU et. al. | arxiv-cs.CV | 2021-04-01 |
| 79 | Detecting Over/under-translation Errors for Determining Adequacy in Human Translations Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a novel approach to detecting over and under translations (OT/UT) as part of adequacy error checks in translation evaluation. |
Prabhakar Gupta; Ridha Juneja; Anil Nelakanti; Tamojit Chatterjee; | arxiv-cs.CL | 2021-04-01 |
| 80 | Low-Resource Neural Machine Translation for Southern African Languages Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Motivated by this challenge we compare zero-shot learning, transfer learning and multilingual learning on three Bantu languages (Shona, isiXhosa and isiZulu) and English. |
Evander Nyoni; Bruce A. Bassett; | arxiv-cs.CL | 2021-04-01 |
| 81 | Sampling and Filtering of Neural Machine Translation Distillation Data Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: The highest-scoring hypothesis of the teacher model is commonly used to train a new model (student). |
Vilém Zouhar; | arxiv-cs.CL | 2021-04-01 |
| 82 | Leveraging Neural Machine Translation for Word Alignment Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We compare this to the extraction of alignment scores from attention. |
Vilém Zouhar; Daria Pylypenko; | arxiv-cs.CL | 2021-03-31 |
| 83 | Domain-specific MT for Low-resource Languages: The Case of Bambara-French Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we address the issue of domain-specific MT for Bambara, an under-resourced Mande language spoken in Mali. We present the first domain-specific parallel dataset for MT of Bambara into and from French. |
Allahsera Auguste Tapo; Michael Leventhal; Sarah Luger; Christopher M. Homan; Marcos Zampieri; | arxiv-cs.CL | 2021-03-31 |
| 84 | Zero-Shot Language Transfer Vs Iterative Back Translation for Unsupervised Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work focuses on comparing different solutions for machine translation on low resource language pairs, namely, with zero-shot transfer learning and unsupervised machine translation. |
Aviral Joshi; Chengzhi Huang; Har Simrat Singh; | arxiv-cs.CL | 2021-03-31 |
| 85 | Divide and Rule: Training Context-Aware Multi-Encoder Translation Models with Little Resources Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we show that training these parameters takes large amount of data, since the contextual training signal is sparse. |
Lorenzo Lupo; Marco Dinarelli; Laurent Besacier; | arxiv-cs.CL | 2021-03-31 |
| 86 | Evaluating The Morphosyntactic Well-formedness of Generated Texts Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose L’AMBRE — a metric to evaluate the morphosyntactic well-formedness of text using its dependency parse and morphosyntactic rules of the language. |
ADITHYA PRATAPA et. al. | arxiv-cs.CL | 2021-03-30 |
| 87 | An Exploration of Data Augmentation Techniques for Improving English to Tigrinya Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present a case study of Tigrinya where we investigate several back-translation methods to generate synthetic source sentences. |
Lidia Kidane; Sachin Kumar; Yulia Tsvetkov; | arxiv-cs.CL | 2021-03-30 |
| 88 | Autocorrect in The Process of Translation — Multi-task Learning Improves Dialogue Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we conduct a deep analysis of a dialogue corpus and summarize three major issues on dialogue translation, including pronoun dropping (\droppro), punctuation dropping (\droppun), and typos (\typo). To properly evaluate the performance, we propose a manually annotated dataset with 1,931 Chinese-English parallel utterances from 300 dialogues as a benchmark testbed for dialogue translation. |
Tao Wang; Chengqi Zhao; Mingxuan Wang; Lei Li; Deyi Xiong; | arxiv-cs.CL | 2021-03-30 |
| 89 | Unsupervised Machine Translation On Dravidian Languages Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we focus on unsupervised translation between English and Kannada, a low resource Dravidian language. |
Sai Koneru; Danni Liu; Jan Niehues; | arxiv-cs.CL | 2021-03-29 |
| 90 | English-Twi Parallel Corpus for Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a parallel machine translation training corpus for English and Akuapem Twi of 25,421 sentence pairs. |
PAUL AZUNRE et. al. | arxiv-cs.CL | 2021-03-29 |
| 91 | Contextual Text Embeddings for Twi Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: The specific contribution of this research work is the development of several pretrained transformer language models for the Akuapem and Asante dialects of Twi, paving the way for advances in application areas such as Named Entity Recognition (NER), Neural Machine Translation (NMT), Sentiment Analysis (SA) and Part-of-Speech (POS) tagging. |
PAUL AZUNRE et. al. | arxiv-cs.CL | 2021-03-29 |
| 92 | PENELOPIE: Enabling Open Information Extraction for The Greek Language Through Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we present our submission for the EACL 2021 SRW; a methodology that aims at bridging the gap between high and low-resource languages in the context of Open Information Extraction, showcasing it on the Greek language. |
Dimitris Papadopoulos; Nikolaos Papadakis; Nikolaos Matsatsinis; | arxiv-cs.CL | 2021-03-28 |
| 93 | Mask Attention Networks: Rethinking and Strengthen Transformer Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a novel understanding of SAN and FFN as Mask Attention Networks (MANs) and show that they are two special cases of MANs with static mask matrices. |
ZHIHAO FAN et. al. | arxiv-cs.CL | 2021-03-25 |
| 94 | Pruning-then-Expanding Model for Domain Adaptation of Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address these three problems, we propose a method of divide and conquer which is based on the importance of neurons or parameters in the translation model. |
Shuhao Gu; Yang Feng; Wanying Xie; | arxiv-cs.CL | 2021-03-25 |
| 95 | Low-Resource Machine Translation for Low-Resource Languages: Leveraging Comparable Data, Code-Switching and Compute Resources Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a simple and scalable method to improve unsupervised NMT, showing how adding comparable data mined using a bilingual dictionary along with modest additional compute resource to train the model can significantly improve its performance. |
Garry Kuwanto; Afra Feyza Akyürek; Isidora Chara Tourni; Siyang Li; Derry Wijaya; | arxiv-cs.CL | 2021-03-24 |
| 96 | Repairing Pronouns in Translation with BERT-Based Post-Editing Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We investigate the severity of this pronoun issue, showing that (1) in some domains, pronoun choice can account for more than half of a NMT systems’ errors, and (2) pronouns have a disproportionately large impact on perceived translation quality. |
Reid Pryzant; | arxiv-cs.CL | 2021-03-23 |
| 97 | The NLP Cookbook: Modern Recipes for Transformer Based Deep Learning Architectures Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we summarize and examine the current state-of-the-art (SOTA) NLP models that have been employed for numerous NLP tasks for optimal performance and efficiency. |
Sushant Singh; Ausif Mahmood; | arxiv-cs.CL | 2021-03-23 |
| 98 | Monolingual and Parallel Corpora for Kangri Low Resource Language Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we present the dataset of Himachali low resource endangered language, Kangri (ISO 639-3xnr) listed in the United Nations Educational, Scientific and Cultural Organization (UNESCO). |
Shweta Chauhan; Shefali Saxena; Philemon Daniel; | arxiv-cs.CL | 2021-03-22 |
| 99 | BlonD: An Automatic Evaluation Metric for Document-level MachineTranslation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address these problems, we propose an automatic metric BlonD for document-level machine translation evaluation. |
YUCHEN JIANG et. al. | arxiv-cs.CL | 2021-03-22 |
| 100 | Non-Autoregressive Translation By Learning Target Categorical Codes Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose CNAT, which learns implicitly categorical codes as latent variables into the non-autoregressive decoding. |
YU BAO et. al. | arxiv-cs.CL | 2021-03-21 |
| 101 | The Effectiveness of Morphology-aware Segmentation in Low-Resource Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper evaluates the performance of several modern subword segmentation methods in a low-resource neural machine translation setting. |
Jonne Sälevä; Constantine Lignos; | arxiv-cs.CL | 2021-03-20 |
| 102 | Dependency Graph-to-String Statistical Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present graph-based translation models which translate source graphs into target strings. |
Liangyou Li; Andy Way; Qun Liu; | arxiv-cs.CL | 2021-03-20 |
| 103 | Congolese Swahili Machine Translation for Humanitarian Response Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we describe our efforts to make a bidirectional Congolese Swahili (SWC) to French (FRA) neural machine translation system with the motivation of improving humanitarian translation workflows. For training, we created a 25,302-sentence general domain parallel corpus and combined it with publicly available data. |
Alp Öktem; Eric DeLuca; Rodrigue Bashizi; Eric Paquin; Grace Tang; | arxiv-cs.CL | 2021-03-19 |
| 104 | Token-wise Curriculum Learning for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address such limitation, we propose a novel token-wise curriculum learning approach that creates sufficient amounts of easy samples. |
CHEN LIANG et. al. | arxiv-cs.CL | 2021-03-19 |
| 105 | Improving The Lexical Ability of Pretrained Language Models for Unsupervised Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we enhance the bilingual masked language model pretraining with lexical-level information by using type-level cross-lingual subword embeddings. |
Alexandra Chronopoulou; Dario Stojanovski; Alexander Fraser; | arxiv-cs.CL | 2021-03-18 |
| 106 | Smoothing and Shrinking The Sparse Seq2Seq Search Space Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we show that entmax-based models effectively solve the cat got your tongue problem, removing a major source of model error for neural machine translation. |
Ben Peters; André F. T. Martins; | arxiv-cs.CL | 2021-03-18 |
| 107 | Gumbel-Attention for Multi-modal Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a novel Gumbel-Attention for multi-modal machine translation, which selects the text-related parts of the image features. |
Pengbo Liu; Hailong Cao; Tiejun Zhao; | arxiv-cs.CL | 2021-03-16 |
| 108 | MENYO-20k: A Multi-domain English-Yorùbá Corpus for Machine Translation and Domain Adaptation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present MENYO-20k, the first multi-domain parallel corpus for the low-resource Yor\`ub\’a–English (yo–en) language pair with standardized train-test splits for benchmarking. We provide several neural MT (NMT) benchmarks on this dataset and compare to the performance of popular pre-trained (massively multilingual) MT models, showing that, in almost all cases, our simple benchmarks outperform the pre-trained MT models. |
DAVID I. ADELANI et. al. | arxiv-cs.CL | 2021-03-15 |
| 109 | Towards The Evaluation of Simultaneous Speech Translation from A Communicative Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present the results of an experiment aimed at evaluating the quality of a simultaneous speech translation engine by comparing it to the performance of professional interpreters. |
claudio Fantinuoli; Bianca Prandi; | arxiv-cs.CL | 2021-03-15 |
| 110 | Crowdsourced Phrase-Based Tokenization for Low-Resourced Neural Machine Translation: The Case of Fon Language Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, using Fon language as a case study, we revisit standard tokenization methods and introduce Word-Expressions-Based (WEB) tokenization, a human-involved super-words tokenization strategy to create a better representative vocabulary for training. |
Bonaventure F. P. Dossou; Chris C. Emezue; | arxiv-cs.CL | 2021-03-14 |
| 111 | OkwuGbé: End-to-End Speech Recognition for Fon and Igbo Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: With the growing awareness and effort to include more low-resourced languages in NLP research, African languages have recently been a major subject of research in machine translation, and other text-based areas of NLP. |
Bonaventure F. P. Dossou; Chris C. Emezue; | arxiv-cs.CL | 2021-03-13 |
| 112 | Improving Translation Robustness with Visual Cues and Error Correction Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we introduce the idea of visual context to improve translation robustness against noisy texts. |
Zhenhao Li; Marek Rei; Lucia Specia; | arxiv-cs.CL | 2021-03-12 |
| 113 | Learning Policies for Multilingual Training of Neural Machine Translation Systems Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose two simple search based curricula — orderings of the multilingual training data — which help improve translation performance in conjunction with existing techniques such as fine-tuning. |
Gaurav Kumar; Philipp Koehn; Sanjeev Khudanpur; | arxiv-cs.CL | 2021-03-11 |
| 114 | Learning Feature Weights Using Reward Modeling for Denoising Parallel Corpora Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work presents an alternative approach which learns weights for multiple sentence-level features. |
Gaurav Kumar; Philipp Koehn; Sanjeev Khudanpur; | arxiv-cs.CL | 2021-03-11 |
| 115 | Unsupervised Transfer Learning in Multilingual Neural Machine Translation with Cross-Lingual Word Embeddings Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work we look into adding a new language to a multilingual NMT system in an unsupervised fashion. |
Carlos Mullov; Ngoc-Quan Pham; Alexander Waibel; | arxiv-cs.CL | 2021-03-11 |
| 116 | Bilingual Dictionary-based Language Model Pretraining for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To alleviate the need for expensive parallel corpora by TLM, in this work, we incorporate the translation information from dictionaries into the pretraining process and propose a novel Bilingual Dictionary-based Language Model (BDLM). |
Yusen Lin; Jiayong Lin; Shuaicheng Zhang; Haoying Dai; | arxiv-cs.CL | 2021-03-11 |
| 117 | Towards Continual Learning for Multilingual Machine Translation Via Vocabulary Substitution Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a straightforward vocabulary adaptation scheme to extend the language capacity of multilingual machine translation models, paving the way towards efficient continual learning for multilingual machine translation. |
Xavier Garcia; Noah Constant; Ankur P. Parikh; Orhan Firat; | arxiv-cs.CL | 2021-03-11 |
| 118 | Self-Learning for Zero Shot Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work proposes a novel zero-shot NMT modeling approach that learns without the now-standard assumption of a pivot language sharing parallel data with the zero-shot source and target languages. |
Surafel M. Lakew; Matteo Negri; Marco Turchi; | arxiv-cs.CL | 2021-03-10 |
| 119 | Weather GAN: Multi-Domain Weather Translation Using Generative Adversarial Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, a new task is proposed, namely, weather translation, which refers to transferring weather conditions of the image from one category to another. |
Xuelong Li; Kai Kou; Bin Zhao; | arxiv-cs.CV | 2021-03-09 |
| 120 | Translating The Unseen? Yoruba-English MT in Low-Resource, Morphologically-Unmarked Settings Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we perform fine-grained analysis on how an SMT system compares with two NMT systems (BiLSTM and Transformer) when translating bare nouns in Yor\`ub\’a into English. |
Ife Adebara; Muhammad Abdul-Mageed; Miikka Silfverberg; | arxiv-cs.CL | 2021-03-06 |
| 121 | Hierarchical Transformer for Multilingual Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we test this idea using the Transformer architecture and show that despite the success in previous work there are problems inherent to training such hierarchical models. |
Albina Khusainova; Adil Khan; Adín Ramírez Rivera; Vitaly Romanov; | arxiv-cs.CL | 2021-03-05 |
| 122 | An Empirical Analysis of Phrase-based and Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We consider phrase reordering models from phrase-based MT systems to investigate which words from inside of a phrase have the biggest impact on defining the phrase reordering behavior. |
Hamidreza Ghader; | arxiv-cs.CL | 2021-03-04 |
| 123 | Meta-Curriculum Learning for Domain Adaptation in Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we aim to alleviate this issue by proposing a novel meta-curriculum learning for domain adaptation in NMT. |
Runzhe Zhan; Xuebo Liu; Derek F. Wong; Lidia S. Chao; | arxiv-cs.CL | 2021-03-03 |
| 124 | Random Feature Attention Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose RFA, a linear time and space attention that uses random feature methods to approximate the softmax function, and explore its application in transformers. |
HAO PENG et. al. | arxiv-cs.CL | 2021-03-02 |
| 125 | Query Rewriting Via Cycle-Consistent Translation for E-Commerce Search Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a novel deep neural network based approach to query rewriting, in order to tackle this problem. |
YIMING QIU et. al. | arxiv-cs.IR | 2021-03-01 |
| 126 | CURE: Code-Aware Neural Machine Translation for Automatic Program Repair Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose CURE, a new NMT-based APR technique with three major novelties. |
Nan Jiang; Thibaud Lutellier; Lin Tan; | arxiv-cs.SE | 2021-02-26 |
| 127 | Gradient-guided Loss Masking for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore strategies that dynamically optimize data usage during the training process using the model’s gradients on a small set of clean data. |
Xinyi Wang; Ankur Bapna; Melvin Johnson; Orhan Firat; | arxiv-cs.CL | 2021-02-26 |
| 128 | Multichannel LSTM-CNN for Telugu Technical Domain Identification Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we proposed the Multichannel LSTM-CNN methodology for Technical Domain Identification for Telugu. |
Sunil Gundapu; Radhika Mamidi; | arxiv-cs.CL | 2021-02-24 |
| 129 | Exploring Supervised and Unsupervised Rewards in Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose to address this problem by making models less reliant on such metrics in two ways: (a) with an entropy-regularised RL method that does not only maximise a reward function but also explore the action space to avoid peaky distributions; (b) with a novel RL method that explores a dynamic unsupervised reward function to balance between exploration and exploitation. |
Julia Ive; Zixu Wang; Marina Fomicheva; Lucia Specia; | arxiv-cs.CL | 2021-02-22 |
| 130 | Exploiting Multimodal Reinforcement Learning for Simultaneous Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: 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. |
JULIA IVE et. al. | arxiv-cs.CL | 2021-02-22 |
| 131 | Understanding and Enhancing The Use of Context for Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this thesis, we focus on understanding certain potentials of contexts in neural models and design augmentation models to benefit from them. |
Marzieh Fadaee; | arxiv-cs.CL | 2021-02-20 |
| 132 | Multi-Domain Adaptation in Neural Machine Translation Through Multidimensional Tagging Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We describe and empirically evaluate multidimensional tagging (MDT), a simple yet effective method for passing sentence-level information to the model. |
Emmanouil Stergiadis; Satendra Kumar; Fedor Kovalev; Pavel Levin; | arxiv-cs.CL | 2021-02-19 |
| 133 | CDA: A Cost Efficient Content-based Multilingual Web Document Aligner Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a Content-based Document Alignment approach (CDA), an efficient method to align multilingual web documents based on content in creating parallel training data for machine translation (MT) systems operating at the industrial level. Besides, we created two web-scale datasets to examine the robustness of CDA in an industrial setting involving up to 28 languages and millions of documents. |
Thuy Vu; Alessandro Moschitti; | arxiv-cs.CL | 2021-02-19 |
| 134 | Machine Translation Customization Via Automatic Training Data Selection from The Web Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: 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. |
Thuy Vu; Alessandro Moschitti; | arxiv-cs.CL | 2021-02-19 |
| 135 | Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To this end, we propose parameterizing hypercomplex multiplications, allowing models to learn multiplication rules from data regardless of whether such rules are predefined. |
ASTON ZHANG et. al. | arxiv-cs.LG | 2021-02-17 |
| 136 | Sparsely Factored Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: 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. |
Noe Casas; Jose A. R. Fonollosa; Marta R. Costa-jussà; | arxiv-cs.CL | 2021-02-17 |
| 137 | Evolving GAN Formulations for Higher Quality Image Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper presents a new technique called TaylorGAN for improving GANs by discovering customized loss functions for each of its two networks. |
Santiago Gonzalez; Mohak Kant; Risto Miikkulainen; | arxiv-cs.NE | 2021-02-17 |
| 138 | Non-Autoregressive Text Generation with Pre-trained Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we show that BERT can be employed as the backbone of a NAG model to greatly improve performance. |
YIXUAN SU et. al. | arxiv-cs.CL | 2021-02-16 |
| 139 | Meta Back-translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a novel method to generate pseudo-parallel data from a pre-trained back-translation model. |
Hieu Pham; Xinyi Wang; Yiming Yang; Graham Neubig; | arxiv-cs.CL | 2021-02-15 |
| 140 | Crowdsourcing Parallel Corpus for English-Oromo Neural Machine Translation Using Community Engagement Platform Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: The paper deals with implementing a translation of English to Afaan Oromo and vice versa using Neural Machine Translation. |
SISAY CHALA et. al. | arxiv-cs.AI | 2021-02-15 |
| 141 | MAPGN: MAsked Pointer-Generator Network for Sequence-to-sequence Pre-training Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper presents a self-supervised learning method for pointer-generator networks to improve spoken-text normalization. |
MANA IHORI et. al. | arxiv-cs.CL | 2021-02-15 |
| 142 | Continuous Learning in Neural Machine Translation Using Bilingual Dictionaries Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we proposed an evaluation framework to assess the ability of neural machine translation to continuously learn new phrases. |
Jan Niehues; | arxiv-cs.CL | 2021-02-12 |
| 143 | Improving Zero-shot Neural Machine Translation on Language-specific Encoders-Decoders Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we study zero-shot translation using language-specific encoders-decoders. |
JUNWEI LIAO et. al. | arxiv-cs.CL | 2021-02-12 |
| 144 | On Automatic Parsing of Log Records Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Instead, we explore the possibility of automating the parsing task by employing machine translation (MT). |
Jared Rand; Andriy Miranskyy; | arxiv-cs.SE | 2021-02-11 |
| 145 | Learning Unsupervised Cross-domain Image-to-Image Translation Using A Shared Discriminator Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a new method that uses a single shared discriminator between the two GANs, which improves the overall efficacy. |
Rajiv Kumar; Rishabh Dabral; G. Sivakumar; | arxiv-cs.CV | 2021-02-09 |
| 146 | Quality Estimation Without Human-labeled Data Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: As an alternative, we propose a technique that does not rely on examples from human-annotators and instead uses synthetic training data. |
YI-LIN TUAN et. al. | arxiv-cs.CL | 2021-02-08 |
| 147 | Understanding Pre-Editing for Black-Box Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To elicit such understanding, we extensively investigated human pre-editing practices. |
Rei Miyata; Atsushi Fujita; | arxiv-cs.CL | 2021-02-04 |
| 148 | One Size Does Not Fit All: Finding The Optimal N-gram Sizes for FastText Models Across Languages Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In our work, we train English, German, Czech, and Italian fastText language models on Wikipedia, and we optimize the subword sizes on the English, German, Czech, and Italian word analogy tasks. |
Vít Novotný; Eniafe Festus Ayetiran; Dávid Lupták; Michal Štefánik; Petr Sojka; | arxiv-cs.CL | 2021-02-04 |
| 149 | Studying The Usage of Text-To-Text Transfer Transformer to Support Code-Related Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we empirically investigate how the T5 model performs when pre-trained and fine-tuned to support code-related tasks. |
ANTONIO MASTROPAOLO et. al. | arxiv-cs.SE | 2021-02-03 |
| 150 | Two Demonstrations of The Machine Translation Applications to Historical Documents Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present our demonstration of two machine translation applications to historical documents. |
Miguel Domingo; Francisco Casacuberta; | arxiv-cs.CL | 2021-02-02 |
| 151 | Machine Translationese: Effects of Algorithmic Bias on Linguistic Complexity in Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we go beyond the study of gender in MT and investigate how bias amplification might affect language in a broader sense. |
Eva Vanmassenhove; Dimitar Shterionov; Matthew Gwilliam; | arxiv-cs.CL | 2021-01-30 |
| 152 | Synthesizing Monolingual Data for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a new method to generate large synthetic parallel data leveraging very small monolingual data in a specific domain. |
Benjamin Marie; Atsushi Fujita; | arxiv-cs.CL | 2021-01-29 |
| 153 | Machine Translation, Sentiment Analysis, Text Similarity, Topic Modelling, and Tweets: Understanding Social Media Usage Among Police and Gendarmerie Organizations Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, it was aimed to investigate the use of social media by the gendarmerie and police organizations operating in Turkey (Jandarma – Polis), Italy (Carabinieri – Polizia), France (Gendarmerie – Police) and Spain (Guardia Civil – Polic\’ia), and the extent to which they can be effective on the followers, by comparatively examining their activity on twitter. |
Emre Cihan Ates; Erkan Bostanci; Mehmet Serdar Guzel; | arxiv-cs.CY | 2021-01-29 |
| 154 | Transition Based Graph Decoder for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a general Transformer-based approach for tree and graph decoding based on generating a sequence of transitions, inspired by a similar approach that uses RNNs by Dyer (2016). |
Leshem Choshen; Omri Abend; | arxiv-cs.CL | 2021-01-29 |
| 155 | Neural Machine Translation, Corpus and Frugality IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: In machine translation field, in both academia and industry, there is a growing interest in increasingly powerful systems, using corpora of several hundred million to several … |
Raoul Blin; | arxiv-cs.CL | 2021-01-26 |
| 156 | A Comparison of Approaches to Document-level Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper presents a systematic comparison of selected approaches from the literature on two benchmarks for which document-level phenomena evaluation suites exist. |
Zhiyi Ma; Sergey Edunov; Michael Auli; | arxiv-cs.CL | 2021-01-26 |
| 157 | Cross-lingual Visual Pre-training for Multimodal Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we combine these two approaches to learn visually-grounded cross-lingual representations. |
OZAN CAGLAYAN et. al. | arxiv-cs.CL | 2021-01-25 |
| 158 | Facilitating Terminology Translation with Target Lemma Annotations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: 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. |
Toms Bergmanis; Mārcis Pinnis; | arxiv-cs.CL | 2021-01-25 |
| 159 | Fast Sequence Generation with Multi-Agent Reinforcement Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a simple and efficient model for Non-Autoregressive sequence Generation (NAG) with a novel training paradigm: Counterfactuals-critical Multi-Agent Learning (CMAL). |
Longteng Guo; Jing Liu; Xinxin Zhu; Hanqing Lu; | arxiv-cs.CL | 2021-01-24 |
| 160 | Towards Natural Language Question Answering Over Earth Observation Linked Data Using Attention-based Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper seeks to study and analyze the use of RNN-based neural machine translation with attention for transforming natural language questions into GeoSPARQL queries. |
Abhishek V. Potnis; Rajat C. Shinde; Surya S. Durbha; | arxiv-cs.CL | 2021-01-23 |
| 161 | GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding IF:3 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we demonstrate conditional computation as a remedy to the above mentioned impediments, and demonstrate its efficacy and utility. |
DMITRY LEPIKHIN et. al. | iclr | 2021-01-21 |
| 162 | Enriching Non-Autoregressive Transformer with Syntactic and SemanticStructures for Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we claim that the syntactic and semantic structures among natural language are critical for non-autoregressive machine translation and can further improve the performance. |
Ye Liu; Yao Wan; Jian-Guo Zhang; Wenting Zhao; Philip S. Yu; | arxiv-cs.CL | 2021-01-21 |
| 163 | Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we attempt to peek into the black-box of multilingual optimization through the lens of loss function geometry. |
Zirui Wang; Yulia Tsvetkov; Orhan Firat; Yuan Cao; | iclr | 2021-01-21 |
| 164 | Nearest Neighbor Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We augment the decoder of a pre-trained machine translation model with a nearest neighbor classifier, substantially improving performance in the single language-pair, multilingual and domain adaptation settings, without any additional training. |
Urvashi Khandelwal; Angela Fan; Dan Jurafsky; Luke Zettlemoyer; Mike Lewis; | iclr | 2021-01-21 |
| 165 | On Learning Universal Representations Across Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we extend pre-trained language models to learn universal representations among multiple languages, and show the effectiveness on cross-lingual understanding and generation. |
XIANGPENG WEI et. al. | iclr | 2021-01-21 |
| 166 | Text Generation By Learning from Off-Policy Demonstrations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose GOLD (generation by off-policy learning from demonstrations): an easy-to-optimize algorithm that learns from the off-policy demonstrations by importance weighting. |
Richard Yuanzhe Pang; He He; | iclr | 2021-01-21 |
| 167 | Parameterization of Hypercomplex Multiplications Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a new parameterization of hypercomplex multiplications for architectural flexibility and effectiveness. |
ASTON ZHANG et. al. | iclr | 2021-01-21 |
| 168 | Disambiguating Symbolic Expressions in Informal Documents Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose the task of \emph{disambiguating} symbolic expressions in informal STEM documents in the form of \LaTeX files — that is, determining their precise semantics and abstract syntax tree — as a neural machine translation task. |
Dennis M�ller; Cezary Kaliszyk; | iclr | 2021-01-21 |
| 169 | Share Or Not? Learning to Schedule Language-Specific Capacity for Multilingual Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We investigate and improve parameter-sharing strategies in multilingual Transformers by utilizing conditional computation. |
Biao Zhang; Ankur Bapna; Rico Sennrich; Orhan Firat; | iclr | 2021-01-21 |
| 170 | Deep Encoder, Shallow Decoder: Reevaluating Non-autoregressive Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We show that the speed disadvantage for autoregressive baselines to evaluate non-autoregressive machine translation is overestimated in three aspects: suboptimal layer allocation, insufficient speed measurement, and lack of knowledge distillation. |
Jungo Kasai; Nikolaos Pappas; Hao Peng; James Cross; Noah Smith; | iclr | 2021-01-21 |
| 171 | Learning Better Structured Representations Using Low-rank Adaptive Label Smoothing Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose an extension of label smoothing which improves generalization performance by adapting to the structure present in label space of structured prediction tasks. |
Asish Ghoshal; Xilun Chen; Sonal Gupta; Luke Zettlemoyer; Yashar Mehdad; | iclr | 2021-01-21 |
| 172 | Random Feature Attention Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a random-feature-based attention that scales linearly in sequence length, and performs on par with strong transformer baselines on language modeling and machine translation. |
HAO PENG et. al. | iclr | 2021-01-21 |
| 173 | Uncertainty Estimation in Autoregressive Structured Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: A Deep Investigation of Ensemble-based Uncertainty Estimation for Autoregressive ASR and NMT models. |
Andrey Malinin; Mark Gales; | iclr | 2021-01-21 |
| 174 | Teach Me How to Label: Labeling Functions from Natural Language with Text-to-text Transformers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: A recent approach to deal with the above issue proposes using natural language explanations instead of labeling individual data points, thereby increasing human annotators’ efficiency as well as decreasing costs substantially. |
Yannis Papanikolaou; | arxiv-cs.CL | 2021-01-18 |
| 175 | GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work introduces GENIE, an extensible human evaluation leaderboard, which brings the ease of leaderboards to text generation tasks. We introduce several datasets in English to GENIE, representing four core challenges in text generation: machine translation, summarization, commonsense reasoning, and machine comprehension. |
DANIEL KHASHABI et. al. | arxiv-cs.CL | 2021-01-16 |
| 176 | The Impact of Post-editing and Machine Translation on Creativity and Reading Experience IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: 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). |
Ana Guerberof Arenas; Antonio Toral; | arxiv-cs.CL | 2021-01-15 |
| 177 | Empirical Evaluation of Supervision Signals for Style Transfer Models IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work we empirically compare the dominant optimization paradigms which provide supervision signals during training: backtranslation, adversarial training and reinforcement learning. |
Yevgeniy Puzikov; Simoes Stanley; Iryna Gurevych; Immanuel Schweizer; | arxiv-cs.CL | 2021-01-15 |
| 178 | Recent Advances in Video Question Answering: A Review of Datasets and Methods IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this survey, we review a number of methods and datasets for the task of VQA. |
Devshree Patel; Ratnam Parikh; Yesha Shastri; | arxiv-cs.CV | 2021-01-14 |
| 179 | Structured Prediction As Translation Between Augmented Natural Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a new framework, Translation between Augmented Natural Languages (TANL), to solve many structured prediction language tasks including joint entity and relation extraction, nested named entity recognition, relation classification, semantic role labeling, event extraction, coreference resolution, and dialogue state tracking. |
GIOVANNI PAOLINI et. al. | arxiv-cs.LG | 2021-01-14 |
| 180 | Context- and Sequence-Aware Convolutional Recurrent Encoder for Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Existing models use recurrent neural networks to construct both the encoder and decoder modules. |
Ritam Mallick; Seba Susan; Vaibhaw Agrawal; Rizul Garg; Prateek Rawal; | arxiv-cs.CL | 2021-01-11 |
| 181 | A Novel Word Sense Disambiguation Approach Using WordNet Knowledge Graph IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper presents a novel knowledge-based word sense disambiguation algorithm, namely Sequential Contextual Similarity Matrix Multiplication (SCSMM). |
Mohannad AlMousa; Rachid Benlamri; Richard Khoury; | arxiv-cs.CL | 2021-01-08 |
| 182 | Decoding Time Lexical Domain Adaptationfor Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our methods are computationally simpler and faster than alternative approaches, and show a moderate success on low-resource settings with explicit out of domain test sets. |
Nikolay Bogoychev; Pinzhen Chen; | arxiv-cs.CL | 2021-01-02 |
| 183 | Zero-shot Learning By Generating Task-specific Adapters IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address these issues, we introduce Hypter, a framework that improves zero-shot transferability by training a hypernetwork to generate task-specific adapters from task descriptions. |
Qinyuan Ye; Xiang Ren; | arxiv-cs.CL | 2021-01-02 |
| 184 | Subformer: Exploring Weight Sharing For Parameter Efficiency In Generative Transformers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Inspired by the success of parameter-sharing in pretrained deep contextualized word representation encoders, we explore parameter-sharing methods in Transformers, with a specific focus on encoder-decoder models for sequence-to-sequence tasks such as neural machine translation. |
Machel Reid; Edison Marrese-Taylor; Yutaka Matsuo; | arxiv-cs.CL | 2021-01-01 |
| 185 | Neural Machine Translation: A Review Of Methods, Resources, And Tools IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: Machine translation (MT) is an important sub-field of natural language processing that aims to translate natural languages using computers. In recent years, end-to-end neural … |
ZHIXING TAN et. al. | arxiv-cs.CL | 2020-12-31 |
| 186 | VOLT: Improving Vocabularization Via Optimal Transport For Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we find an exciting relation between an information-theoretic feature and BLEU scores. |
Jingjing Xu; Hao Zhou; Chun Gan; Zaixiang Zheng; Lei Li; | arxiv-cs.CL | 2020-12-31 |
| 187 | Fully Non-autoregressive Neural Machine Translation: Tricks Of The Trade IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we target on closing the performance gap while maintaining the latency advantage. |
Jiatao Gu; Xiang Kong; | arxiv-cs.CL | 2020-12-31 |
| 188 | Exploring Monolingual Data For Neural Machine Translation With Knowledge Distillation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We explore two types of monolingual data that can be included in knowledge distillation training for neural machine translation (NMT). |
Alham Fikri Aji; Kenneth Heafield; | arxiv-cs.CL | 2020-12-31 |
| 189 | XLM-T: Scaling Up Multilingual Machine Translation With Pretrained Cross-lingual Transformer Encoders IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, inspired by the recent success of language model pre-training, we present XLM-T, which initializes the model with an off-the-shelf pretrained cross-lingual Transformer encoder and fine-tunes it with multilingual parallel data. |
SHUMING MA et. al. | arxiv-cs.CL | 2020-12-31 |
| 190 | FDMT: A Benchmark Dataset For Fine-grained Domain Adaptation In Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To motivate a wide investigation in such settings, we present a real-world fine-grained domain adaptation task in machine translation (FDMT). |
WENHAO ZHU et. al. | arxiv-cs.CL | 2020-12-31 |
| 191 | Revisiting Robust Neural Machine Translation: A Transformer Case Study IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, we introduce a novel data-driven technique to incorporate noise during training. |
Peyman Passban; Puneeth S. M. Saladi; Qun Liu; | arxiv-cs.CL | 2020-12-31 |
| 192 | Directed Beam Search: Plug-and-Play Lexically Constrained Language Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose Directed Beam Search (DBS), a plug-and-play method for lexically constrained language generation. |
Damian Pascual; Beni Egressy; Florian Bolli; Roger Wattenhofer; | arxiv-cs.CL | 2020-12-30 |
| 193 | Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, our main contribution is to take a step further in understanding EncoderFusion. |
XUEBO LIU et. al. | arxiv-cs.CL | 2020-12-29 |
| 194 | Neural Text Generation With Artificial Negative Examples IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose to suppress an arbitrary type of errors by training the text generation model in a reinforcement learning framework, where we use a trainable reward function that is capable of discriminating between references and sentences containing the targeted type of errors. |
Keisuke Shirai; Kazuma Hashimoto; Akiko Eriguchi; Takashi Ninomiya; Shinsuke Mori; | arxiv-cs.CL | 2020-12-28 |
| 195 | Towards Fully Automated Manga Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we make the following four contributions that establishes the foundation of manga translation research. Third, we created a new benchmark to evaluate manga translation. |
Ryota Hinami; Shonosuke Ishiwatari; Kazuhiko Yasuda; Yusuke Matsui; | arxiv-cs.CL | 2020-12-28 |
| 196 | Understanding and Improving Lexical Choice in Non-Autoregressive Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, we empirically show that as a side effect of this training, the lexical choice errors on low-frequency words are propagated to the NAT model from the teacher model. |
LIANG DING et. al. | arxiv-cs.CL | 2020-12-28 |
| 197 | Learning Light-Weight Translation Models From Deep Transformer IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we take a natural step towards learning strong but light-weight NMT systems. |
BEI LI et. al. | arxiv-cs.CL | 2020-12-27 |
| 198 | SG-Net: Syntax Guided Transformer for Language Representation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose using syntax to guide the text modeling by incorporating explicit syntactic constraints into attention mechanisms for better linguistically motivated word representations. |
ZHUOSHENG ZHANG et. al. | arxiv-cs.CL | 2020-12-27 |
| 199 | Why Neural Machine Translation Prefers Empty Outputs IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We investigate why neural machine translation (NMT) systems assign high probability to empty translations. |
Xing Shi; Yijun Xiao; Kevin Knight; | arxiv-cs.CL | 2020-12-24 |
| 200 | Gender Bias In Multilingual Neural Machine Translation: The Architecture Matters IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, and from an algorithmic perspective, we explore if the chosen architecture, when trained with the same data, influences the gender bias accuracy. |
MARTA R. COSTA-JUSSÀ et. al. | arxiv-cs.CL | 2020-12-24 |
| 201 | Multi-Head Self-Attention With Role-Guided Masks IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a method to guide the attention heads towards roles identified in prior work as important. |
DONGSHENG WANG et. al. | arxiv-cs.CL | 2020-12-22 |
| 202 | Domain Adaptation Of NMT Models For English-Hindi Machine Translation Task At AdapMT ICON 2020 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes the neural machine translation systems for the English-Hindi language presented in AdapMT Shared Task ICON 2020. |
Ramchandra Joshi; Rushabh Karnavat; Kaustubh Jirapure; Raviraj Joshi; | arxiv-cs.CL | 2020-12-22 |
| 203 | Subword Sampling For Low Resource Word Alignment IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose subword sampling-based alignment of text units. |
Ehsaneddin Asgari; Masoud Jalili Sabet; Philipp Dufter; Christopher Ringlstetter; Hinrich Schütze; | arxiv-cs.CL | 2020-12-21 |
| 204 | Learning To Predict Vulnerabilities From Vulnerability-Fixes: A Machine Translation Approach IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To deal with this issue, we develop a prediction method using the encoder-decoder framework of machine translation that automatically learns the latent features (context, patterns, etc.) of code that are linked with vulnerabilities. |
AAYUSH GARG et. al. | arxiv-cs.CR | 2020-12-21 |
| 205 | Finding Sparse Structures for Domain Specific Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To mitigate the issue, we propose Prune-Tune, a novel domain adaptation method via gradual pruning. |
Jianze Liang; Chengqi Zhao; Mingxuan Wang; Xipeng Qiu; Lei Li; | arxiv-cs.CL | 2020-12-18 |
| 206 | Named Entity Recognition In The Legal Domain Using A Pointer Generator Network IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To overcome the challenges of our noisy training data, e.g. text extraction errors and/or typos and unknown label indices, we formulate the NER task as a text-to-text sequence generation task and train a pointer generator network to generate the entities in the document rather than label them. |
Stavroula Skylaki; Ali Oskooei; Omar Bari; Nadja Herger; Zac Kriegman; | arxiv-cs.CL | 2020-12-17 |
| 207 | Exploring Fluent Query Reformulations With Text-to-Text Transformers And Reinforcement Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We explore methods to generate these query reformulations by training reformulators using text-to-text transformers and apply policy-based reinforcement learning algorithms to further encourage reward learning. |
Jerry Zikun Chen; Shi Yu; Haoran Wang; | arxiv-cs.CL | 2020-12-17 |
| 208 | Improving Multilingual Neural Machine Translation For Low-Resource Languages: French-, English- Vietnamese IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper proposes two simple strategies to address the rare word issue in multilingual MT systems for two low-resource language pairs: French-Vietnamese and English-Vietnamese. |
Thi-Vinh Ngo; Phuong-Thai Nguyen; Thanh-Le Ha; Khac-Quy Dinh; Le-Minh Nguyen; | arxiv-cs.CL | 2020-12-15 |
| 209 | Exploring Transfer Learning For End-to-End Spoken Language Understanding IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose an E2E system that is designed to jointly train on multiple speech-to-text tasks, such as ASR (speech-transcription) and SLU (speech-hypothesis), and text-to-text tasks, such as NLU (text-hypothesis). We will soon release the audio data collected for the TOP dataset for future research. |
SUBENDHU RONGALI et. al. | arxiv-cs.CL | 2020-12-15 |
| 210 | Modeling Homophone Noise For Robust Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a robust neural machine translation (NMT) framework. |
WENJIE QIN et. al. | arxiv-cs.CL | 2020-12-15 |
| 211 | Contrastive Learning with Adversarial Perturbations for Conditional Text Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose to mitigate the conditional text generation problem by contrasting positive pairs with negative pairs, such that the model is exposed to various valid or incorrect perturbations of the inputs, for improved generalization. |
Seanie Lee; Dong Bok Lee; Sung Ju Hwang; | arxiv-cs.CL | 2020-12-14 |
| 212 | Mask-Align: Self-Supervised Neural Word Alignment IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose Mask-Align, a self-supervised model specifically designed for the word alignment task. |
Chi Chen; Maosong Sun; Yang Liu; | arxiv-cs.CL | 2020-12-13 |
| 213 | MSVD-Turkish: A Comprehensive Multimodal Dataset For Integrated Vision And Language Research In Turkish IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we target Turkish, a morphologically rich and agglutinative language that has very different properties compared to English. |
BEGUM CITAMAK et. al. | arxiv-cs.CV | 2020-12-13 |
| 214 | Document-aligned Japanese-English Conversation Parallel Corpus IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address the first issue, we present a document-aligned Japanese-English conversation corpus, including balanced, high-quality business conversation data for tuning and testing. |
Matīss Rikters; Ryokan Ri; Tong Li; Toshiaki Nakazawa; | arxiv-cs.CL | 2020-12-11 |
| 215 | Automatic Standardization Of Colloquial Persian IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes a simple and effective standardization approach based on sequence-to-sequence translation. Moreover, we annotate a publicly available evaluation data consisting of 1912 sentences from a diverse set of domains. |
Mohammad Sadegh Rasooli; Farzane Bakhtyari; Fatemeh Shafiei; Mahsa Ravanbakhsh; Chris Callison-Burch; | arxiv-cs.CL | 2020-12-10 |
| 216 | Image-Graph-Image Translation Via Auto-Encoding IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work presents the first convolutional neural network that learns an image-to-graph translation task without needing external supervision. |
Chenyang Lu; Gijs Dubbelman; | arxiv-cs.CV | 2020-12-10 |
| 217 | Multilingual Transfer Learning For QA Using Translation As Data Augmentation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we explore strategies that improve cross-lingual transfer by bringing the multilingual embeddings closer in the semantic space. |
Mihaela Bornea; Lin Pan; Sara Rosenthal; Radu Florian; Avirup Sil; | arxiv-cs.CL | 2020-12-10 |
| 218 | Exploring Pair-Wise NMT For Indian Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we address the task of improving pair-wise machine translation for specific low resource Indian languages. |
KARTHEEK AKELLA et. al. | arxiv-cs.CL | 2020-12-10 |
| 219 | Approches Quantitatives De L’analyse Des Pr{é}dictions En Traduction Automatique Neuronale (TAN) IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: As part of a larger project on optimal learning conditions in neural machine translation, we investigate characteristic training phases of translation engines. |
Maria Zimina-Poirot; Nicolas Ballier; Jean-Baptiste Yunès; | arxiv-cs.CL | 2020-12-10 |
| 220 | Towards Annotation-Free Evaluation Of Cross-Lingual Image Captioning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In order to save the precious human resource from re-writing reference sentences per target language, in this paper we make a brave attempt towards annotation-free evaluation of cross-lingual image captioning. |
Aozhu Chen; Xinyi Huang; Hailan Lin; Xirong Li; | arxiv-cs.CV | 2020-12-09 |
| 221 | On Knowledge Distillation For Direct Speech Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we compare the different solutions to distill knowledge in a sequence-to-sequence task like ST. Moreover, we analyze eventual drawbacks of this approach and how to alleviate them maintaining the benefits in terms of translation quality. |
Marco Gaido; Mattia A. Di Gangi; Matteo Negri; Marco Turchi; | arxiv-cs.CL | 2020-12-09 |
| 222 | Breeding Gender-aware Direct Speech Translation Systems IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Going beyond speech signals, in this paper we compare different approaches to inform direct ST models about the speaker’s gender and test their ability to handle gender translation from English into Italian and French. |
Marco Gaido; Beatrice Savoldi; Luisa Bentivogli; Matteo Negri; Marco Turchi; | arxiv-cs.CL | 2020-12-09 |
| 223 | Rewriter-Evaluator Architecture for Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address this issue, we present a novel architecture, Rewriter-Evaluator. |
Yangming Li; Kaisheng Yao; | arxiv-cs.CL | 2020-12-09 |
| 224 | Globetrotter: Unsupervised Multilingual Translation From Visual Alignment IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a framework that instead uses the visual modality to align multiple languages, using images as the bridge between them. |
Dídac Surís; Dave Epstein; Carl Vondrick; | arxiv-cs.CL | 2020-12-08 |
| 225 | Document Graph For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Previous works have shown that contextual information can improve the performance of neural machine translation (NMT). |
Mingzhou Xu; Liangyou Li; Derek. F. Wong; Qun Liu; Lidia S. Chao; | arxiv-cs.CL | 2020-12-07 |
| 226 | Reciprocal Supervised Learning Improves Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we revisit the utilization of multiple diverse models and present a simple yet effective approach named Reciprocal-Supervised Learning (RSL). |
MINKAI XU et. al. | arxiv-cs.CL | 2020-12-05 |
| 227 | A Benchmark Dataset For Understandable Medical Language Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we introduce MedLane — a new human-annotated Medical Language translation dataset, to align professional medical sentences with layperson-understandable expressions. |
JUNYU LUO et. al. | arxiv-cs.CL | 2020-12-04 |
| 228 | Quality Estimation & Interpretability for Code Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we attempt to estimate the quality of source code translations built on top of the TransCoder model. |
MAYANK AGARWAL et. al. | arxiv-cs.SE | 2020-12-04 |
| 229 | SemMT: A Semantic-based Testing Approach For Machine Translation Systems IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, in this paper, we propose SemMT, an automatic testing approach for machine translation systems based on semantic similarity checking. |
Jialun Cao; Meiziniu Li; Yeting Li; Ming Wen; Shing-Chi Cheung; | arxiv-cs.SE | 2020-12-03 |
| 230 | A Simple And Effective Approach To Robust Unsupervised Bilingual Dictionary Induction IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we show that this failure results from the gap between the actual initialization performance and the minimum initialization performance for the self-learning to succeed. |
YANYANG LI et. al. | arxiv-cs.CL | 2020-11-30 |
| 231 | Dynamic Curriculum Learning For Low-Resource Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In particular, we propose a dynamic curriculum learning (DCL) method to reorder training samples in training. |
CHEN XU et. al. | arxiv-cs.CL | 2020-11-30 |
| 232 | Machine Translation Of Novels In The Age Of Transformer IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this chapter we build a machine translation (MT) system tailored to the literary domain, specifically to novels, based on the state-of-the-art architecture in neural MT (NMT), the Transformer (Vaswani et al., 2017), for the translation direction English-to-Catalan. |
Antonio Toral; Antoni Oliver; Pau Ribas Ballestín; | arxiv-cs.CL | 2020-11-30 |
| 233 | A Comprehensive Review On Recent Methods And Challenges Of Video Description IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we report a comprehensive survey on the phases of video description approaches, the dataset for video description, evaluation metrics, open competitions for motivating the research on the video description, open challenges in this field, and future research directions. |
Alok Singh; Thoudam Doren Singh; Sivaji Bandyopadhyay; | arxiv-cs.CV | 2020-11-30 |
| 234 | Using Multiple Subwords To Improve English-Esperanto Automated Literary Translation Quality IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To that end, in this paper we propose employing the same parallel sentences multiple times, only changing the way the words are split each time. |
Alberto Poncelas; Jan Buts; James Hadley; Andy Way; | arxiv-cs.CL | 2020-11-28 |
| 235 | Unsupervised Word Translation Pairing Using Refinement Based Point Set Registration IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper proposes BioSpere, a novel framework for unsupervised mapping of bi-lingual word embeddings onto a shared vector space, by combining adversarial initialization and refinement procedure with point set registration algorithm used in image processing. |
Silviu Oprea; Sourav Dutta; Haytham Assem; | arxiv-cs.CL | 2020-11-26 |
| 236 | Decoding And Diversity In Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we characterize distributional differences between generated and real translations, examining the cost in diversity paid for the BLEU scores enjoyed by NMT. |
Nicholas Roberts; Davis Liang; Graham Neubig; Zachary C. Lipton; | arxiv-cs.CL | 2020-11-26 |
| 237 | A Panoramic Survey Of Natural Language Processing In The Arab World IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: NLP researchers pride themselves on developing language independent models and tools that can be applied to all human languages, e.g. machine translation systems can be built for a variety of languages using the same basic mechanisms and models. |
KAREEM DARWISH et. al. | arxiv-cs.CL | 2020-11-25 |
| 238 | Two-Way Neural Machine Translation: A Proof Of Concept For Bidirectional Translation Modeling Using A Two-Dimensional Grid IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper proposes to build a single end-to-end bidirectional translation model using a two-dimensional grid, where the left-to-right decoding generates source-to-target, and the bottom-to-up decoding creates target-to-source output. |
Parnia Bahar; Christopher Brix; Hermann Ney; | arxiv-cs.CL | 2020-11-24 |
| 239 | The Ubiqus English-Inuktitut System For WMT20 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes Ubiqus’ submission to the WMT20 English-Inuktitut shared news translation task. |
François Hernandez; Vincent Nguyen; | arxiv-cs.CL | 2020-11-18 |
| 240 | Master Thesis: Neural Sign Language Translation By Learning Tokenization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this thesis, we propose a multitask learning based method to improve Neural Sign Language Translation (NSLT) consisting of two parts, a tokenization layer and Neural Machine Translation (NMT). |
Alptekin Orbay; | arxiv-cs.CL | 2020-11-18 |
| 241 | Unsupervised Translation Of Programming Languages IF:3 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose to leverage recent approaches in unsupervised machine translation to train a fully unsupervised neural transcompiler. |
Baptiste Roziere; Marie-Anne Lachaux; Lowik Chanussot; Guillaume Lample; | nips | 2020-11-17 |
| 242 | Deep Transformers With Latent Depth IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a probabilistic framework to automatically learn which layer(s) to use by learning the posterior distributions of layer selection. |
Xian Li; Asa Cooper Stickland; Yuqing Tang; Xiang Kong; | nips | 2020-11-17 |
| 243 | Data Diversification: A Simple Strategy For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce Data Diversification: a simple but effective strategy to boost neural machine translation (NMT) performance. |
Xuan-Phi Nguyen; Shafiq Joty; Kui Wu; Ai Ti Aw; | nips | 2020-11-17 |
| 244 | Sparse And Continuous Attention Mechanisms IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper expands that work in two directions: first, we extend alpha-entmax to continuous domains, revealing a link with Tsallis statistics and deformed exponential families. Second, we introduce continuous-domain attention mechanisms, deriving efficient gradient backpropagation algorithms for alpha in {1,2}. |
ANDR� MARTINS et. al. | nips | 2020-11-17 |
| 245 | Is Normalization Indispensable For Training Deep Neural Network? IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we study what would happen when normalization layers are removed from the network, and show how to train deep neural networks without normalization layers and without performance degradation. |
Jie Shao; Kai Hu; Changhu Wang; Xiangyang Xue; Bhiksha Raj; | nips | 2020-11-17 |
| 246 | Cascaded Text Generation With Markov Transformers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work proposes an autoregressive model with sub-linear parallel time generation. |
Yuntian Deng; Alexander Rush; | nips | 2020-11-17 |
| 247 | All Word Embeddings From One Embedding IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, to reduce the total number of parameters, the embeddings for all words are represented by transforming a shared embedding. |
Sho Takase; Sosuke Kobayashi; | nips | 2020-11-17 |
| 248 | Compositional Generalization Via Neural-Symbolic Stack Machines IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To tackle this issue, we propose the Neural-Symbolic Stack Machine (NeSS). |
Xinyun Chen; Chen Liang; Adams Wei Yu; Dawn Song; Denny Zhou; | nips | 2020-11-17 |
| 249 | Incorporating BERT Into Parallel Sequence Decoding With Adapters IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose to address this problem by taking two different BERT models as the encoder and decoder respectively, and fine-tuning them by introducing simple and lightweight adapter modules, which are inserted between BERT layers and tuned on the task-specific dataset. |
JUNLIANG GUO et. al. | nips | 2020-11-17 |
| 250 | Cross-lingual Retrieval For Iterative Self-Supervised Training IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we found that the cross-lingual alignment can be further improved by training seq2seq models on sentence pairs mined using their own encoder outputs. |
Chau Tran; Yuqing Tang; Xian Li; Jiatao Gu; | nips | 2020-11-17 |
| 251 | Facebook AI’s WMT20 News Translation Task Submission IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes Facebook AI’s submission to WMT20 shared news translation task. |
PENG-JEN CHEN et. al. | arxiv-cs.CL | 2020-11-16 |
| 252 | A Hybrid Approach for Improved Low Resource Neural Machine Translation Using Monolingual Data Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work, therefore, proposes a novel approach that enables both the backward and forward models to benefit from the monolingual target data through a hybrid of self-learning and back-translation respectively. |
Idris Abdulmumin; Bashir Shehu Galadanci; Abubakar Isa; Ismaila Idris Sinan; | arxiv-cs.CL | 2020-11-14 |
| 253 | EDITOR: An Edit-Based Transformer with Repositioning for Neural Machine Translation with Soft Lexical Constraints IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce an Edit-Based Transformer with Repositioning (EDITOR), which makes sequence generation flexible by seamlessly allowing users to specify preferences in output lexical choice. |
Weijia Xu; Marine Carpuat; | arxiv-cs.CL | 2020-11-13 |
| 254 | Language Models Not Just For Pre-training: Fast Online Neural Noisy Channel Modeling IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We address this issue by introducing efficient approximations to make inference with the noisy channel approach as fast as strong ensembles while increasing accuracy. |
Shruti Bhosale; Kyra Yee; Sergey Edunov; Michael Auli; | arxiv-cs.CL | 2020-11-13 |
| 255 | BLEU Might Be Guilty But References Are Not Innocent IF:3 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We study different methods to collect references and compare their value in automated evaluation by reporting correlation with human evaluation for a variety of systems and metrics. |
Markus Freitag; David Grangier; Isaac Caswell; | emnlp | 2020-11-12 |
| 256 | Distilling Multiple Domains For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a framework for training a single multi-domain neural machine translation model that is able to translate several domains without increasing inference time or memory usage. |
Anna Currey; Prashant Mathur; Georgiana Dinu; | emnlp | 2020-11-12 |
| 257 | Monolingual Adapters For Zero-Shot Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a novel adapter layer formalism for adapting multilingual models. |
Jerin Philip; Alexandre Berard; Matthias Gallé; Laurent Besacier; | emnlp | 2020-11-12 |
| 258 | Losing Heads In The Lottery: Pruning Transformer Attention In Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we apply the lottery ticket hypothesis to prune heads in the early stages of training. |
Maximiliana Behnke; Kenneth Heafield; | emnlp | 2020-11-12 |
| 259 | Generating Diverse Translation From Model Distribution With Dropout IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose to generate diverse translations by deriving a large number of possible models with Bayesian modelling and sampling models from them for inference. |
Xuanfu Wu; Yang Feng; Chenze Shao; | emnlp | 2020-11-12 |
| 260 | Data Rejuvenation: Exploiting Inactive Training Examples For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we explore to identify the inactive training examples which contribute less to the model performance, and show that the existence of inactive examples depends on the data distribution. |
WENXIANG JIAO et. al. | emnlp | 2020-11-12 |
| 261 | COMET: A Neural Framework For MT Evaluation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present COMET, a neural framework for training multilingual machine translation evaluation models which obtains new state-of-the-art levels of correlation with human judgements. |
Ricardo Rei; Craig Stewart; Ana C Farinha; Alon Lavie; | emnlp | 2020-11-12 |
| 262 | Why Skip If You Can Combine: A Simple Knowledge Distillation Technique For Intermediate Layers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we target low-resource settings and evaluate our translation engines for Portuguese?English, Turkish?English, and English?German directions. |
Yimeng Wu; Peyman Passban; Mehdi Rezagholizadeh; Qun Liu; | emnlp | 2020-11-12 |
| 263 | Bridging Linguistic Typology And Multilingual Machine Translation With Multi-View Language Representations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose to fuse both views using singular vector canonical correlation analysis and study what kind of information is induced from each source. |
Arturo Oncevay; Barry Haddow; Alexandra Birch; | emnlp | 2020-11-12 |
| 264 | Reusing A Pretrained Language Model On Languages With Limited Corpora For Unsupervised NMT IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present an effective approach that reuses an LM that is pretrained only on the high-resource language. |
Alexandra Chronopoulou; Dario Stojanovski; Alexander Fraser; | emnlp | 2020-11-12 |
| 265 | Inference-only Sub-character Decomposition Improves Translation Of Unseen Logographic Characters IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We investigate existing ideograph-based sub-character decomposition approaches for Chinese-to-English and Japanese-to-English NMT, for both high-resource and low-resource domains. |
Danielle Saunders; Weston Feely; Bill Byrne; | arxiv-cs.CL | 2020-11-12 |
| 266 | Data And Representation For Turkish Natural Language Inference IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we offer a positive response for natural language inference (NLI) in Turkish. |
Emrah Budur; Rıza Özçelik; Tunga Gungor; Christopher Potts; | emnlp | 2020-11-12 |
| 267 | Can Automatic Post-Editing Improve NMT? IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We hypothesize that APE models have been underperforming in improving NMT translations due to the lack of adequate supervision. To ascertain our hypothesis, we compile a larger corpus of human post-edits of English to German NMT. |
Shamil Chollampatt; Raymond Hendy Susanto; Liling Tan; Ewa Szymanska; | emnlp | 2020-11-12 |
| 268 | Towards Enhancing Faithfulness For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a novel training strategy with a multi-task learning paradigm to build a faithfulness enhanced NMT model (named FEnmt). |
Rongxiang Weng; Heng Yu; Xiangpeng Wei; Weihua Luo; | emnlp | 2020-11-12 |
| 269 | Token-level Adaptive Training For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explored target token-level adaptive objectives based on token frequencies to assign appropriate weights for each target token during training. |
SHUHAO GU et. al. | emnlp | 2020-11-12 |
| 270 | Inference Strategies For Machine Translation With Conditional Masking IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We identify a thresholding strategy that has advantages over the standard mask-predict algorithm, and provide analyses of its behavior on machine translation tasks. |
Julia Kreutzer; George Foster; Colin Cherry; | emnlp | 2020-11-12 |
| 271 | Statistical Power And Translationese In Machine Translation Evaluation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: The term translationese has been used to describe features of translated text, and in this paper, we provide detailed analysis of potential adverse effects of translationese on machine translation evaluation. |
Yvette Graham; Barry Haddow; Philipp Koehn; | emnlp | 2020-11-12 |
| 272 | Pre-tokenization Of Multi-word Expressions In Cross-lingual Word Embeddings IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a simple method for word translation of MWEs to and from English in ten languages: we first compile lists of MWEs in each language and then tokenize the MWEs as single tokens before training word embeddings. |
NAOKI OTANI et. al. | emnlp | 2020-11-12 |
| 273 | LAReQA: Language-Agnostic Answer Retrieval From A Multilingual Pool IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present LAReQA, a challenging new benchmark for language-agnostic answer retrieval from a multilingual candidate pool. |
UMA ROY et. al. | emnlp | 2020-11-12 |
| 274 | Shallow-to-Deep Training For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we investigate the behavior of a well-tuned deep Transformer system. |
BEI LI et. al. | emnlp | 2020-11-12 |
| 275 | Alignment-free Cross-lingual Semantic Role Labeling IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a cross-lingual SRL model which only requires annotations in a source language and access to raw text in the form of a parallel corpus. |
Rui Cai; Mirella Lapata; | emnlp | 2020-11-12 |
| 276 | Multi-task Learning For Multilingual Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a multi-task learning (MTL) framework that jointly trains the model with the translation task on bitext data and two denoising tasks on the monolingual data. |
Yiren Wang; ChengXiang Zhai; Hany Hassan; | emnlp | 2020-11-12 |
| 277 | Improving AMR Parsing With Sequence-to-Sequence Pre-training IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we focus on sequence-to-sequence (seq2seq) AMR parsing and propose a seq2seq pre-training approach to build pre-trained models in both single and joint way on three relevant tasks, i.e., machine translation, syntactic parsing, and AMR parsing itself. |
Dongqin Xu; Junhui Li; Muhua Zhu; Min Zhang; Guodong Zhou; | emnlp | 2020-11-12 |
| 278 | Self-Paced Learning For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We ameliorate this procedure with a more flexible manner by proposing self-paced learning, where NMT model is allowed to 1) automatically quantify the learning confidence over training examples; and 2) flexibly govern its learning via regulating the loss in each iteration step. |
YU WAN et. al. | emnlp | 2020-11-12 |
| 279 | ChrEn: Cherokee-English Machine Translation For Endangered Language Revitalization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To help save this endangered language, we introduce ChrEn, a Cherokee-English parallel dataset, to facilitate machine translation research between Cherokee and English. |
Shiyue Zhang; Benjamin Frey; Mohit Bansal; | emnlp | 2020-11-12 |
| 280 | If Beam Search Is The Answer, What Was The Question? IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We frame beam search as the exact solution to a different decoding objective in order to gain insights into why high probability under a model alone may not indicate adequacy. |
Clara Meister; Ryan Cotterell; Tim Vieira; | emnlp | 2020-11-12 |
| 281 | Automatic Machine Translation Evaluation In Many Languages Via Zero-Shot Paraphrasing IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose training the paraphraser as a multilingual NMT system, treating paraphrasing as a zero-shot translation task (e.g., Czech to Czech). |
Brian Thompson; Matt Post; | emnlp | 2020-11-12 |
| 282 | Iterative Refinement In The Continuous Space For Non-Autoregressive Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose an efficient inference procedure for non-autoregressive machine translation that iteratively refines translation purely in the continuous space. |
Jason Lee; Raphael Shu; Kyunghyun Cho; | emnlp | 2020-11-12 |
| 283 | Simulated Multiple Reference Training Improves Low-resource Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce Simulated Multiple Reference Training (SMRT), a novel MT training method that approximates the full space of possible translations by sampling a paraphrase of the reference sentence from a paraphraser and training the MT model to predict the paraphraser’s distribution over possible tokens. |
Huda Khayrallah; Brian Thompson; Matt Post; Philipp Koehn; | emnlp | 2020-11-12 |
| 284 | Coding Textual Inputs Boosts The Accuracy Of Neural Networks IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: As alternatives to a text representation, we introduce Soundex, MetaPhone, NYSIIS, logogram to NLP, and develop fixed-output-length coding and its extension using Huffman coding. |
Abdul Rafae Khan; Jia Xu; Weiwei Sun; | emnlp | 2020-11-12 |
| 285 | Detecting Word Sense Disambiguation Biases In Machine Translation For Model-Agnostic Adversarial Attacks IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a method for the prediction of disambiguation errors based on statistical data properties, demonstrating its effectiveness across several domains and model types. |
Denis Emelin; Ivan Titov; Rico Sennrich; | emnlp | 2020-11-12 |
| 286 | Improving Word Sense Disambiguation With Translations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a novel approach that improves the performance of a base WSD system using machine translation. |
Yixing Luan; Bradley Hauer; Lili Mou; Grzegorz Kondrak; | emnlp | 2020-11-12 |
| 287 | Dynamic Data Selection And Weighting For Iterative Back-Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we provide insights into this commonly used approach and generalize it to a dynamic curriculum learning strategy, which is applied to iterative back-translation models. |
Zi-Yi Dou; Antonios Anastasopoulos; Graham Neubig; | emnlp | 2020-11-12 |
| 288 | Uncertainty-Aware Semantic Augmentation For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address this problem, we propose uncertainty-aware semantic augmentation, which explicitly captures the universal semantic information among multiple semantically-equivalent source sentences and enhances the hidden representations with this information for better translations. |
XIANGPENG WEI et. al. | emnlp | 2020-11-12 |
| 289 | TeaForN: Teacher-Forcing With N-grams IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our proposed method, Teacher-Forcing with N-grams (TeaForN), addresses both these problems directly, through the use of a stack of N decoders trained to decode along a secondary time axis that allows model-parameter updates based on N prediction steps. |
Sebastian Goodman; Nan Ding; Radu Soricut; | emnlp | 2020-11-12 |
| 290 | On The Sparsity Of Neural Machine Translation Models IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In response to this problem, we empirically investigate whether the redundant parameters can be reused to achieve better performance. |
Yong Wang; Longyue Wang; Victor Li; Zhaopeng Tu; | emnlp | 2020-11-12 |
| 291 | Language Model Prior For Low-Resource Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a novel approach to incorporate a LM as prior in a neural translation model (TM). |
Christos Baziotis; Barry Haddow; Alexandra Birch; | emnlp | 2020-11-12 |
| 292 | An Empirical Study Of Generation Order For Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present an empirical study of generation order for machine translation. |
William Chan; Mitchell Stern; Jamie Kiros; Jakob Uszkoreit; | emnlp | 2020-11-12 |
| 293 | Dynamic Context Selection For Document-level Neural Machine Translation Via Reinforcement Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address this problem, we propose an effective approach to select dynamic context so that the document-level translation model can utilize the more useful selected context sentences to produce better translations. |
Xiaomian Kang; Yang Zhao; Jiajun Zhang; Chengqing Zong; | emnlp | 2020-11-12 |
| 294 | Localizing Open-Ontology QA Semantic Parsers In A Day Using Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose Semantic Parser Localizer (SPL), a toolkit that leverages Neural Machine Translation (NMT) systems to localize a semantic parser for a new language. |
Mehrad Moradshahi; Giovanni Campagna; Sina Semnani; Silei Xu; Monica Lam; | emnlp | 2020-11-12 |
| 295 | Revisiting Modularized Multilingual NMT To Meet Industrial Demands IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, we revisit the multilingual neural machine translation model that only share modules among the same languages (M2) as a practical alternative to 1-1 to satisfy industrial requirements. |
Sungwon Lyu; Bokyung Son; Kichang Yang; Jaekyoung Bae; | emnlp | 2020-11-12 |
| 296 | Self-Induced Curriculum Learning In Self-Supervised Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, we provide an in-depth analysis of the sampling choices the SSNMT model makes during training. |
Dana Ruiter; Josef van Genabith; Cristina España-Bonet; | emnlp | 2020-11-12 |
| 297 | Improving Text Generation With Student-Forcing Optimal Transport IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To reduce this gap between training and testing, we propose using optimal transport (OT) to match the sequences generated in these two modes. |
JIANQIAO LI et. al. | emnlp | 2020-11-12 |
| 298 | Long-Short Term Masking Transformer: A Simple But Effective Baseline For Document-level Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we research extensively the pros and cons of the standard transformer in document-level translation, and find that the auto-regressive property can simultaneously bring both the advantage of the consistency and the disadvantage of error accumulation. |
Pei Zhang; Boxing Chen; Niyu Ge; Kai Fan; | emnlp | 2020-11-12 |
| 299 | Effectively Pretraining A Speech Translation Decoder With Machine Translation Data IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we will show that by using an adversarial regularizer, we can bring the encoder representations of the ASR and NMT tasks closer even though they are in different modalities, and how this helps us effectively use a pretrained NMT decoder for speech translation. |
Ashkan Alinejad; Anoop Sarkar; | emnlp | 2020-11-12 |
| 300 | Simultaneous Machine Translation With Visual Context IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we seek to understand whether the addition of visual information can compensate for the missing source context. |
OZAN CAGLAYAN et. al. | emnlp | 2020-11-12 |
| 301 | X-SRL: A Parallel Cross-Lingual Semantic Role Labeling Dataset IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work we propose a method to automatically construct an SRL corpus that is parallel in four languages: English, French, German, Spanish, with unified predicate and role annotations that are fully comparable across languages. |
Angel Daza; Anette Frank; | emnlp | 2020-11-12 |
| 302 | Multi-Unit Transformers For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose the Multi-Unit Transformer (MUTE) , which aim to promote the expressiveness of the Transformer by introducing diverse and complementary units. |
Jianhao Yan; Fandong Meng; Jie Zhou; | emnlp | 2020-11-12 |
| 303 | Not Low-Resource Anymore: Aligner Ensembling, Batch Filtering, And New Datasets For Bengali-English Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we build a customized sentence segmenter for Bengali and propose two novel methods for parallel corpus creation on low-resource setups: aligner ensembling and batch filtering. We release the segmenter, parallel corpus, and the evaluation set, thus elevating Bengali from its low-resource status. |
TAHMID HASAN et. al. | emnlp | 2020-11-12 |
| 304 | Incomplete Utterance Rewriting As Semantic Segmentation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a novel and extensive approach, which formulates it as a semantic segmentation task. |
Qian Liu; Bei Chen; Jian-Guang Lou; Bin Zhou; Dongmei Zhang; | emnlp | 2020-11-12 |
| 305 | A Streaming Approach For Efficient Batched Beam Search IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose an efficient batching strategy for variable-length decoding on GPU architectures. |
Kevin Yang; Violet Yao; John DeNero; Dan Klein; | emnlp | 2020-11-12 |
| 306 | Pronoun-Targeted Fine-tuning For NMT With Hybrid Losses IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a class of conditional generative-discriminative hybrid losses that we use to fine-tune a trained machine translation model. |
Prathyusha Jwalapuram; Shafiq Joty; Youlin Shen; | emnlp | 2020-11-12 |
| 307 | Detecting Fine-Grained Cross-Lingual Semantic Divergences Without Supervision By Learning To Rank IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a training strategy for multilingual BERT models by learning to rank synthetic divergent examples of varying granularity. |
Eleftheria Briakou; Marine Carpuat; | emnlp | 2020-11-12 |
| 308 | Direct Segmentation Models For Streaming Speech Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work proposes novel segmentation models for streaming ST that incorporate not only textual, but also acoustic information to decide when the ASR output is split into a chunk. |
JAVIER IRANZO-SÁNCHEZ et. al. | emnlp | 2020-11-12 |
| 309 | PyMT5: Multi-mode Translation Of Natural Language And Python Code With Transformers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present an analysis and modeling effort of a large-scale parallel corpus of 26 million Python methods and 7.7 million method-docstring pairs, demonstrating that for docstring and method generation, PyMT5 outperforms similarly-sized auto-regressive language models (GPT2) which were English pre-trained or randomly initialized. |
Colin Clement; Dawn Drain; Jonathan Timcheck; Alexey Svyatkovskiy; Neel Sundaresan; | emnlp | 2020-11-12 |
| 310 | CSP:Code-Switching Pre-training For Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper proposes a new pre-training method, called Code-Switching Pre-training (CSP for short) for Neural Machine Translation (NMT). |
Zhen Yang; Bojie Hu; Ambyera Han; Shen Huang; Qi Ju; | emnlp | 2020-11-12 |
| 311 | Towards Reasonably-Sized Character-Level Transformer NMT By Finetuning Subword Systems Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We show that by initially training a subword model and then finetuning it on characters, we can obtain a neural machine translation model that works at the character level without requiring token segmentation. |
Jindřich Libovický; Alexander Fraser; | emnlp | 2020-11-12 |
| 312 | Evaluating Curriculum Learning Strategies In Neural Combinatorial Optimization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we focus on designing a curriculum learning-based training procedure that can help existing architectures achieve competitive performance on a large range of problem sizes simultaneously. |
Michal Lisicki; Arash Afkanpour; Graham W. Taylor; | arxiv-cs.LG | 2020-11-11 |
| 313 | Zero-Pair Image To Image Translation Using Domain Conditional Normalization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose an approach based on domain conditional normalization (DCN) for zero-pair image-to-image translation, i.e., translating between two domains which have no paired training data available but each have paired training data with a third domain. |
Samarth Shukla; Andrés Romero; Luc Van Gool; Radu Timofte; | arxiv-cs.CV | 2020-11-11 |
| 314 | Fixed Encoder Self-Attention Patterns In Transformer-Based Machine Translation IF:3 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose to replace all but one attention head of each encoder layer with simple fixed – non-learnable – attentive patterns that are solely based on position and do not require any external knowledge. |
Alessandro Raganato; Yves Scherrer; Jörg Tiedemann; | emnlp | 2020-11-10 |
| 315 | SimAlign: High Quality Word Alignments Without Parallel Training Data Using Static And Contextualized Embeddings IF:3 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose word alignment methods that require no parallel data. |
Masoud Jalili Sabet; Philipp Dufter; François Yvon; Hinrich Schütze; | emnlp | 2020-11-10 |
| 316 | A Semi-supervised Approach To Generate The Code-Mixed Text Using Pre-trained Encoder And Transfer Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To circumvent the data scarcity issue, we propose an effective deep learning approach for automatically generating the code-mixed text from English to multiple languages without any parallel data. |
Deepak Gupta; Asif Ekbal; Pushpak Bhattacharyya; | emnlp | 2020-11-10 |
| 317 | Adversarial Subword Regularization For Robust Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present adversarial subword regularization (ADVSR) to study whether gradient signals during training can be a substitute criterion for exposing diverse subword segmentations. |
Jungsoo Park; Mujeen Sung; Jinhyuk Lee; Jaewoo Kang; | emnlp | 2020-11-10 |
| 318 | FELIX: Flexible Text Editing Through Tagging And Insertion IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present FELIX – a flexible text-editing approach for generation, designed to derive maximum benefit from the ideas of decoding with bi-directional contexts and self-supervised pretraining. |
Jonathan Mallinson; Aliaksei Severyn; Eric Malmi; Guillermo Garrido; | emnlp | 2020-11-10 |
| 319 | Using Visual Feature Space As A Pivot Across Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our work aims to leverage visual feature space to pass information across languages. |
Ziyan Yang; Leticia Pinto-Alva; Franck Dernoncourt; Vicente Ordonez; | emnlp | 2020-11-10 |
| 320 | Bootstrapping A Crosslingual Semantic Parser IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We develop a Transformer-based parser combining paraphrases by ensembling attention over multiple encoders and present new versions of ATIS and Overnight in German and Chinese for evaluation. |
Tom Sherborne; Yumo Xu; Mirella Lapata; | emnlp | 2020-11-10 |
| 321 | Participatory Research For Low-resourced Machine Translation: A Case Study In African Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we focus on the task of Machine Translation (MT), that plays a crucial role for information accessibility and communication worldwide. |
WILHELMINA NEKOTO et. al. | emnlp | 2020-11-10 |
| 322 | Factorized Transformer For Multi-Domain Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To mitigate these problems, we propose the Factorized Transformer, which consists of an in-depth factorization of the parameters of an NMT model, namely Transformer in this paper, into two categories: domain-shared ones that encode common cross-domain knowledge and domain-specific ones that are private for each constituent domain. |
Yongchao Deng; Hongfei Yu; Heng Yu; Xiangyu Duan; Weihua Luo; | emnlp | 2020-11-10 |
| 323 | On Long-Tailed Phenomena In Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we quantitatively characterize such long-tailed phenomena at two levels of abstraction, namely, token classification and sequence generation. |
Vikas Raunak; Siddharth Dalmia; Vivek Gupta; Florian Metze; | emnlp | 2020-11-10 |
| 324 | Unsupervised Contrastive Photo-to-Caricature Translation Based On Auto-distortion IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To take both into consideration, we propose an unsupervised contrastive photo-to-caricature translation architecture. |
Yuhe Ding; Xin Ma; Mandi Luo; Aihua Zheng; Ran He; | arxiv-cs.CV | 2020-11-10 |
| 325 | Assessing Human-Parity In Machine Translation On The Segment Level IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To provide insight into these issues we carefully inspect the outputs of top-performing systems in the most recent WMT-19 news translation shared task for all language pairs in which a system either tied or outperformed human translation. |
Yvette Graham; Christian Federmann; Maria Eskevich; Barry Haddow; | emnlp | 2020-11-10 |
| 326 | Vocabulary Adaptation For Domain Adaptation In Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, aiming to solve these vocabulary mismatches in domain adaptation for neural machine translation (NMT), we propose vocabulary adaptation, a simple method for effective fine-tuning that adapts embedding layers in a given pretrained NMT model to the target domain. |
Shoetsu Sato; Jin Sakuma; Naoki Yoshinaga; Masashi Toyoda; Masaru Kitsuregawa; | emnlp | 2020-11-10 |
| 327 | Recursive Top-Down Production For Sentence Generation With Latent Trees IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To this end, we present a dynamic programming algorithm that marginalises over latent binary tree structures with N leaves, allowing us to compute the likelihood of a sequence of N tokens under a latent tree model, which we maximise to train a recursive neural function. |
Shawn Tan; Yikang Shen; Alessandro Sordoni; Aaron Courville; Timothy J. O’Donnell; | emnlp | 2020-11-10 |
| 328 | Reference Language Based Unsupervised Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Further enriching the idea of pivot translation by extending the use of parallel corpora beyond the source-target paradigm, we propose a new reference language-based framework for UNMT, RUNMT, in which the reference language only shares a parallel corpus with the source, but this corpus still indicates a signal clear enough to help the reconstruction training of UNMT through a proposed reference agreement mechanism. |
Zuchao Li; Hai Zhao; Rui Wang; Masao Utiyama; Eiichiro Sumita; | emnlp | 2020-11-10 |
| 329 | Pushing The Limits Of AMR Parsing With Self-Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore different ways in which trained models can be applied to improve AMR parsing performance, including generation of synthetic text and AMR annotations as well as refinement of actions oracle. |
YOUNG-SUK LEE et. al. | emnlp | 2020-11-10 |
| 330 | Neural Machine Translation For Extremely Low-Resource African Languages: A Case Study On Bambara IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present the first parallel data set for machine translation of Bambara into and from English and French and the first benchmark results on machine translation to and from Bambara. |
ALLAHSERA AUGUSTE TAPO et. al. | arxiv-cs.CL | 2020-11-10 |
| 331 | Multi-Agent Mutual Learning At Sentence-Level And Token-Level For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we extend mutual learning to the machine translation task and operate at both the sentence-level and the token-level. |
Baohao Liao; Yingbo Gao; Hermann Ney; | emnlp | 2020-11-10 |
| 332 | From Unsupervised Machine Translation To Adversarial Text Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a self-attention based bilingual adversarial text generator (B-GAN) which can learn to generate text from the encoder representation of an unsupervised neural machine translation system. |
Ahmad Rashid; Alan Do-Omri; Md. Akmal Haidar; Qun Liu; Mehdi Rezagholizadeh; | arxiv-cs.CL | 2020-11-10 |
| 333 | Finding The Optimal Vocabulary Size For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We cast neural machine translation (NMT) as a classification task in an autoregressive setting and analyze the limitations of both classification and autoregression components. |
Thamme Gowda; Jonathan May; | emnlp | 2020-11-10 |
| 334 | Scheduled DropHead: A Regularization Method For Transformer Models IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce DropHead, a structured dropout method specifically designed for regularizing the multi-head attention mechanism which is a key component of transformer. |
Wangchunshu Zhou; Tao Ge; Furu Wei; Ming Zhou; Ke Xu; | emnlp | 2020-11-10 |
| 335 | Multilingual Argument Mining: Datasets And Analysis IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we explore the potential of transfer learning using the multilingual BERT model to address argument mining tasks in non-English languages, based on English datasets and the use of machine translation. |
Orith Toledo-Ronen; Matan Orbach; Yonatan Bilu; Artem Spector; Noam Slonim; | emnlp | 2020-11-10 |
| 336 | Improving Grammatical Error Correction With Machine Translation Pairs IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: 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 (e.g., Chinese to English) of different qualities (i.e., poor and good). |
WANGCHUNSHU ZHOU et. al. | emnlp | 2020-11-10 |
| 337 | Dual Reconstruction: A Unifying Objective For Semi-Supervised Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce a novel dual reconstruction objective that provides a unified view of Iterative Back-Translation and Dual Learning. |
Weijia Xu; Xing Niu; Marine Carpuat; | emnlp | 2020-11-10 |
| 338 | KoBE: Knowledge-Based Machine Translation Evaluation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a simple and effective method for machine translation evaluation which does not require reference translations. |
Zorik Gekhman; Roee Aharoni; Genady Beryozkin; Markus Freitag; Wolfgang Macherey; | emnlp | 2020-11-10 |
| 339 | On Romanization For Model Transfer Between Scripts In Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We explore the benefit of romanization in this scenario. |
Chantal Amrhein; Rico Sennrich; | emnlp | 2020-11-10 |
| 340 | Scene Graph Modification Based On Natural Language Commands IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore the novel problem of graph modification, where the systems need to learn how to update an existing scene graph given a new user’s command. |
XUANLI HE et. al. | emnlp | 2020-11-10 |
| 341 | Transformer-GCRF: Recovering Chinese Dropped Pronouns With General Conditional Random Fields IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a novel framework that combines the strength of Transformer network with General Conditional Random Fields (GCRF) to model the dependencies between pronouns in neighboring utterances. |
JINGXUAN YANG et. al. | emnlp | 2020-11-10 |
| 342 | On The Sub-Layer Functionalities Of Transformer Decoder IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we study how Transformer-based decoders leverage information from the source and target languages – developing a universal probe task to assess how information is propagated through each module of each decoder layer. |
YILIN YANG et. al. | emnlp | 2020-11-10 |
| 343 | On The Potential Of Lexico-logical Alignments For Semantic Parsing To SQL Queries IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose and test two methods: (1) supervised attention; (2) adopting an auxiliary objective of disambiguating references in the input queries to table columns. |
Tianze Shi; Chen Zhao; Jordan Boyd-Graber; Hal Daumé III; Lillian Lee; | emnlp | 2020-11-10 |
| 344 | A Multilingual View Of Unsupervised Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a probabilistic framework for multilingual neural machine translation that encompasses supervised and unsupervised setups, focusing on unsupervised translation. |
Xavier Garcia; Pierre Foret; Thibault Sellam; Ankur Parikh; | emnlp | 2020-11-10 |
| 345 | Automatically Identifying Gender Issues In Machine Translation Using Perturbations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: While previous studies have identified issues using synthetic examples, we develop a novel technique to mine examples from real world data to explore challenges for deployed systems. |
Hila Gonen; Kellie Webster; | emnlp | 2020-11-10 |
| 346 | WikiLingua: A New Benchmark Dataset For Multilingual Abstractive Summarization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce WikiLingua, a large-scale, multilingual dataset for the evaluation of cross-lingual abstractive summarization systems. |
Faisal Ladhak; Esin Durmus; Claire Cardie; Kathleen McKeown; | emnlp | 2020-11-10 |
| 347 | Graph-to-Tree Neural Networks For Learning Structured Input-Output Translation With Applications To Semantic Parsing And Math Word Problem IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a novel Graph-to-Tree Neural Networks, namely Graph2Tree consisting of a graph encoder and a hierarchical tree decoder, that encodes an augmented graph-structured input and decodes a tree-structured output. |
SHUCHENG LI et. al. | emnlp | 2020-11-10 |
| 348 | Training Flexible Depth Model By Multi-Task Learning For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose to use multi-task learning to train a flexible depth model that can adapt to different depth configurations during inference. |
Qiang Wang; Tong Xiao; Jingbo Zhu; | emnlp | 2020-11-10 |
| 349 | Hybrid Emoji-Based Masked Language Models For Zero-Shot Abusive Language Detection IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: More specifically, we propose a hybrid emoji-based Masked Language Model (MLM) to leverage the common information conveyed by emojis across different languages and improve the learned cross-lingual representation of short text messages, with the goal to perform zero- shot abusive language detection. |
Michele Corazza; Stefano Menini; Elena Cabrio; Sara Tonelli; Serena Villata; | emnlp | 2020-11-10 |
| 350 | PharmMT: A Neural Machine Translation Approach To Simplify Prescription Directions IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a novel machine translation-based approach, PharmMT, to automatically and reliably simplify prescription directions into patient-friendly language, thereby significantly reducing pharmacist workload. |
JIAZHAO LI et. al. | emnlp | 2020-11-10 |
| 351 | Active Learning Approaches To Enhancing Neural Machine Translation: An Empirical Study IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore new training frameworks by incorporating active learning into various techniques such as transfer learning and iterative back-translation (IBT) under a limited human translation budget. |
Yuekai Zhao; Haoran Zhang; Shuchang Zhou; Zhihua Zhang; | emnlp | 2020-11-10 |
| 352 | It’s Not A Non-Issue: Negation As A Source Of Error In Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study we focus on negation, a universal, core property of human language that significantly affects the semantics of an utterance. |
Md Mosharaf Hossain; Antonios Anastasopoulos; Eduardo Blanco; Alexis Palmer; | emnlp | 2020-11-10 |
| 353 | Adversarial Grammatical Error Correction IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present an adversarial learning approach to GEC, using the generator-discriminator framework. |
Vipul Raheja; Dimitris Alikaniotis; | emnlp | 2020-11-10 |
| 354 | The Box Is In The Pen: Evaluating Commonsense Reasoning In Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a test suite to evaluate the commonsense reasoning capability of neural machine translation. |
Jie He; Tao Wang; Deyi Xiong; Qun Liu; | emnlp | 2020-11-10 |
| 355 | Improving Word Embedding Factorization For Compression Using Distilled Nonlinear Neural Decomposition IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose Distilled Embedding, an (input/output) embedding compression method based on low-rank matrix decomposition and knowledge distillation. |
Vasileios Lioutas; Ahmad Rashid; Krtin Kumar; Md. Akmal Haidar; Mehdi Rezagholizadeh; | emnlp | 2020-11-10 |
| 356 | Fully Quantized Transformer For Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To this end, we propose FullyQT: an all-inclusive quantization strategy for the Transformer. |
Gabriele Prato; Ella Charlaix; Mehdi Rezagholizadeh; | emnlp | 2020-11-10 |
| 357 | Improving Target-side Lexical Transfer In Multilingual Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we aim to improve the effectiveness of multilingual transfer for NMT models that translate into the LRL, by designing a better decoder word embedding. |
Luyu Gao; Xinyi Wang; Graham Neubig; | emnlp | 2020-11-10 |
| 358 | Converting The Point Of View Of Message Spoken To Virtual Assistants Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We designed a system to allow virtual assistants to take a voice message from one user, convert the point of view of the message, and then deliver the result to its target user. |
GUNHEE LEE et. al. | emnlp | 2020-11-10 |
| 359 | Computer Assisted Translation With Neural Quality Estimation And Auotmatic Post-Editing Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose an end-to-end deep learning framework of the quality estimation and automatic post-editing of the machine translation output. |
KE WANG et. al. | emnlp | 2020-11-10 |
| 360 | Simultaneous Speech-to-Speech Translation System With Neural Incremental ASR, MT, And TTS IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper presents a newly developed, simultaneous neural speech-to-speech translation system and its evaluation. |
KATSUHITO SUDOH et. al. | arxiv-cs.CL | 2020-11-09 |
| 361 | BERT-JAM: Boosting BERT-Enhanced Neural Machine Translation With Joint Attention IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a novel BERT-enhanced NMT model called BERT-JAM which improves upon existing models from two aspects: 1) BERT-JAM uses joint-attention modules to allow the encoder/decoder layers to dynamically allocate attention between different representations, and 2) BERT-JAM allows the encoder/decoder layers to make use of BERT’s intermediate representations by composing them using a gated linear unit (GLU). |
Zhebin Zhang; Sai Wu; Dawei Jiang; Gang Chen; | arxiv-cs.CL | 2020-11-09 |
| 362 | Template Controllable Keywords-to-text Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper proposes a novel neural model for the understudied task of generating text from keywords. |
Abhijit Mishra; Md Faisal Mahbub Chowdhury; Sagar Manohar; Dan Gutfreund; Karthik Sankaranarayanan; | arxiv-cs.AI | 2020-11-07 |
| 363 | AlphaMWE: Construction Of Multilingual Parallel Corpora With MWE Annotations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present the construction of multilingual parallel corpora with annotation of multiword expressions (MWEs). |
Lifeng Han; Gareth Jones; Alan Smeaton; | arxiv-cs.CL | 2020-11-07 |
| 364 | Understanding Pure Character-Based Neural Machine Translation: The Case Of Translating Finnish Into English IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we conduct an investigation into pure character-based models in the case of translating Finnish into English, including exploring the ability to learn word senses and morphological inflections and the attention mechanism. |
Gongbo Tang; Rico Sennrich; Joakim Nivre; | arxiv-cs.CL | 2020-11-06 |
| 365 | An Unsupervised Method For OCR Post-Correction And Spelling Normalisation For Finnish IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We build on previous work on fully automatic unsupervised extraction of parallel data to train a character-based sequence-to-sequence NMT (neural machine translation) model to conduct OCR error correction designed for English, and adapt it to Finnish by proposing solutions that take the rich morphology of the language into account. |
Quan Duong; Mika Hämäläinen; Simon Hengchen; | arxiv-cs.CL | 2020-11-06 |
| 366 | Semi-Supervised Low-Resource Style Transfer Of Indonesian Informal To Formal Language With Iterative Forward-Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we address a style-transfer from informal to formal Indonesian as a low-resource machine translation problem. We build a new dataset of parallel sentences of informal Indonesian and its formal counterpart. |
HARYO AKBARIANTO WIBOWO et. al. | arxiv-cs.CL | 2020-11-06 |
| 367 | Unleashing The Power Of Neural Discourse Parsers — A Context And Structure Aware Approach Using Large Scale Pretraining Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we demonstrate a simple, yet highly accurate discourse parser, incorporating recent contextual language models. |
Grigorii Guz; Patrick Huber; Giuseppe Carenini; | arxiv-cs.CL | 2020-11-06 |
| 368 | Data Augmentation and Terminology Integration for Domain-Specific Sinhala-English-Tamil Statistical Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper focuses on data augmentation techniques where bilingual lexicon terms are expanded based on case-markers with the objective of generating new words, to be used in Statistical machine Translation (SMT). |
Aloka Fernando; Surangika Ranathunga; Gihan Dias; | arxiv-cs.CL | 2020-11-05 |
| 369 | Cross-Lingual Machine Speech Chain For Javanese, Sundanese, Balinese, And Bataks Speech Recognition And Synthesis IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, we focus on developing speech recognition and synthesis for these Indonesian ethnic languages: Javanese, Sundanese, Balinese, and Bataks. |
Sashi Novitasari; Andros Tjandra; Sakriani Sakti; Satoshi Nakamura; | arxiv-cs.CL | 2020-11-04 |
| 370 | MK-SQuIT: Synthesizing Questions Using Iterative Template-filling IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: The aim of this work is to create a framework for synthetically generating question/query pairs with as little human input as possible. We also present an example dataset of 110,000 question/query pairs across four WikiData domains. |
Benjamin A. Spiegel; Vincent Cheong; James E. Kaplan; Anthony Sanchez; | arxiv-cs.CL | 2020-11-04 |
| 371 | Detecting Hallucinated Content In Conditional Neural Sequence Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To better assess the faithfulness of the machine outputs, we propose a new task to predict whether each token in the output sequence is hallucinated conditioned on the source input, and collect new manually annotated evaluation sets for this task. We will release our annotated data and code to support future research. |
CHUNTING ZHOU et. al. | arxiv-cs.CL | 2020-11-04 |
| 372 | Optimizing Transformer For Low-Resource Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: Language pairs with limited amounts of parallel data, also known as low-resource languages, remain a challenge for neural machine translation. While the Transformer model has … |
Ali Araabi; Christof Monz; | arxiv-cs.CL | 2020-11-04 |
| 373 | PheMT: A Phenomenon-wise Dataset For Machine Translation Robustness On User-Generated Contents IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To answer the question, we present a new dataset, PheMT, for evaluating the robustness of MT systems against specific linguistic phenomena in Japanese-English translation. |
RYO FUJII et. al. | arxiv-cs.CL | 2020-11-03 |
| 374 | Subword Segmentation And A Single Bridge Language Affect Zero-Shot Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we investigate zero-shot performance of a multilingual EN$\leftrightarrow${FR,CS,DE,FI} system trained on WMT data. |
Annette Rios; Mathias Müller; Rico Sennrich; | arxiv-cs.CL | 2020-11-03 |
| 375 | Cross-lingual Word Embeddings Beyond Zero-shot Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We explore the transferability of a multilingual neural machine translation model to unseen languages when the transfer is grounded solely on the cross-lingual word embeddings. |
Shifei Chen; Ali Basirat; | arxiv-cs.CL | 2020-11-03 |
| 376 | Layer-Wise Multi-View Learning For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose layer-wise multi-view learning to solve this problem, circumventing the necessity to change the model structure. |
Qiang Wang; Changliang Li; Yue Zhang; Tong Xiao; Jingbo Zhu; | arxiv-cs.CL | 2020-11-03 |
| 377 | The 2020s Political Economy Of Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper explores the hypothesis that the diversity of human languages, right now a barrier to interoperability in communication and trade, will become significantly less of a barrier as machine translation technologies are deployed over the next several years.But this new boundary-breaking technology does not reduce all boundaries equally, and it creates new challenges for the distribution of ideas and thus for innovation and economic growth. |
Steven Weber; | arxiv-cs.CY | 2020-11-02 |
| 378 | Emergent Communication Pretraining For Few-Shot Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: For instance, in this work we measure the influence of communication success and maximum sequence length on downstream performances. |
Yaoyiran Li; Edoardo M. Ponti; Ivan Vulić; Anna Korhonen; | arxiv-cs.CL | 2020-11-02 |
| 379 | Enabling Zero-shot Multilingual Spoken Language Translation With Language-Specific Encoders And Decoders IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our proposed method extends a MultiNMT architecture based on language-specific encoders-decoders to the task of Multilingual SLT (MultiSLT) Our experiments on four different languages show that coupling the speech encoder to the MultiNMT architecture produces similar quality translations compared to a bilingual baseline ($\pm 0.2$ BLEU) while effectively allowing for zero-shot MultiSLT. |
Carlos Escolano; Marta R. Costa-jussà; José A. R. Fonollosa; Carlos Segura; | arxiv-cs.CL | 2020-11-02 |
| 380 | An Empirical Study Of Contextual Data Augmentation For Japanese Zero Anaphora Resolution IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We adopt a state-of-the-art data augmentation method, called the contextual data augmentation (CDA), that generates labeled training instances using a pretrained language model. |
RYUTO KONNO et. al. | arxiv-cs.CL | 2020-11-02 |
| 381 | Investigating Catastrophic Forgetting During Continual Training For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Neural machine translation (NMT) models usually suffer from catastrophic forgetting during continual training where the models tend to gradually forget previously learned knowledge and swing to fit the newly added data which may have a different distribution, e.g. a different domain. |
Shuhao Gu; Yang Feng; | arxiv-cs.CL | 2020-11-01 |
| 382 | A Targeted Attack on Black-Box Neural Machine Translation with Parallel Data Poisoning Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we show that targeted attacks on black-box NMT systems are feasible, based on poisoning a small fraction of their parallel training data. |
CHANG XU et. al. | arxiv-cs.CL | 2020-11-01 |
| 383 | Tilde At WMT 2020: News Task Systems IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes Tilde’s submission to the WMT2020 shared task on news translation for both directions of the English-Polish language pair in both the constrained and the unconstrained tracks. |
Rihards Krišlauks; Mārcis Pinnis; | arxiv-cs.CL | 2020-10-29 |
| 384 | Bridging The Modality Gap For Speech-to-Text Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address these issues, we propose a Speech-to-Text Adaptation for Speech Translation (STAST) model which aims to improve the end-to-end model performance by bridging the modality gap between speech and text. |
Yuchen Liu; Junnan Zhu; Jiajun Zhang; Chengqing Zong; | arxiv-cs.CL | 2020-10-28 |
| 385 | The Volctrans Machine Translation System For WMT20 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes our VolcTrans system on WMT20 shared news translation task. |
LIWEI WU et. al. | arxiv-cs.CL | 2020-10-28 |
| 386 | Exploiting Neural Query Translation Into Cross Lingual Information Retrieval IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we investigate how to exploit neural query translation model into CLIR system. |
Liang Yao; Baosong Yang; Haibo Zhang; Weihua Luo; Boxing Chen; | arxiv-cs.CL | 2020-10-26 |
| 387 | Constraint Translation Candidates: A Bridge Between Neural Query Translation And Cross-lingual Information Retrieval IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a novel approach to alleviate these problems by limiting the open target vocabulary search space of QT to a set of important words mined from search index database. |
TIANCHI BI et. al. | arxiv-cs.CL | 2020-10-26 |
| 388 | Is It Great Or Terrible? Preserving Sentiment In Neural Machine Translation Of Arabic Reviews IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we investigate the challenges involved in translating book reviews from Arabic into English, with particular focus on the errors that lead to incorrect translation of sentiment polarity. |
Hadeel Saadany; Constantin Orasan; | arxiv-cs.CL | 2020-10-26 |
| 389 | Data Troubles In Sentence Level Confidence Estimation For Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: As a side product of the data annotation process necessary for building such models we propose sentence level accuracy $SACC$ as a simple, self-explanatory evaluation metric for quality of translation. |
CIPRIAN CHELBA et. al. | arxiv-cs.CL | 2020-10-26 |
| 390 | The LMU Munich System For The WMT 2020 Unsupervised Machine Translation Shared Task IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes the submission of LMU Munich to the WMT 2020 unsupervised shared task, in two language directions, German<->Upper Sorbian. |
Alexandra Chronopoulou; Dario Stojanovski; Viktor Hangya; Alexander Fraser; | arxiv-cs.CL | 2020-10-25 |
| 391 | Multi-Task Learning With Shared Encoder For Non-Autoregressive Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we hypothesize and empirically verify that AT and NAT encoders capture different linguistic properties and representations of source sentences. |
YONGCHANG HAO et. al. | arxiv-cs.CL | 2020-10-24 |
| 392 | Unsupervised Vision-and-Language Pre-training Without Parallel Images and Captions IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In particular, we propose to conduct “mask-and-predict” pre-training on text-only and image-only corpora and introduce the object tags detected by an object recognition model as anchor points to bridge two modalities. |
LIUNIAN HAROLD LI et. al. | arxiv-cs.CL | 2020-10-24 |
| 393 | Context-aware Decoder For Neural Machine Translation Using A Target-side Document-Level Language Model IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We therefore present a simple method to turn a sentence-level translation model into a context-aware model by incorporating a document-level language model into the decoder. |
Amane Sugiyama; Naoki Yoshinaga; | arxiv-cs.CL | 2020-10-24 |
| 394 | On Learning Text Style Transfer With Direct Rewards IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore training algorithms that instead optimize reward functions that explicitly consider different aspects of the style-transferred outputs. |
Yixin Liu; Graham Neubig; John Wieting; | arxiv-cs.CL | 2020-10-24 |
| 395 | ReadOnce Transformers: Reusable Representations Of Text For Transformers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a transformer-based approach, ReadOnce Transformers, that is trained to build such information-capturing representations of text. |
Shih-Ting Lin; Ashish Sabharwal; Tushar Khot; | arxiv-cs.CL | 2020-10-24 |
| 396 | Anchor-based Bilingual Word Embeddings For Low-Resource Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper proposes a new approach for building BWEs in which the vector space of the high resource source language is used as a starting point for training an embedding space for the low resource target language. |
Tobias Eder; Viktor Hangya; Alexander Fraser; | arxiv-cs.CL | 2020-10-23 |
| 397 | Rapid Domain Adaptation For Machine Translation With Monolingual Data IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose an approach that enables rapid domain adaptation from the perspective of unsupervised translation. |
Mahdis Mahdieh; Mia Xu Chen; Yuan Cao; Orhan Firat; | arxiv-cs.CL | 2020-10-23 |
| 398 | CUNI Systems For The Unsupervised And Very Low Resource Translation Task In WMT20 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper presents a description of CUNI systems submitted to the WMT20 task on unsupervised and very low-resource supervised machine translation between German and Upper Sorbian. |
Ivana Kvapilíková; Tom Kocmi; Ondřej Bojar; | arxiv-cs.CL | 2020-10-22 |
| 399 | XOR QA: Cross-lingual Open-Retrieval Question Answering IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Based on this dataset, we introduce three new tasks that involve cross-lingual document retrieval using multi-lingual and English resources. We construct a large-scale dataset built on questions from TyDi QA lacking same-language answers. |
AKARI ASAI et. al. | arxiv-cs.CL | 2020-10-22 |
| 400 | Not All Parameters Are Born Equal: Attention Is Mostly What You Need IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To gain insight into this, we consider three groups of parameters: embeddings, attention, and feed forward neural network (FFN) layers. |
Nikolay Bogoychev; | arxiv-cs.CL | 2020-10-22 |
| 401 | MT5: A Massively Multilingual Pre-trained Text-to-text Transformer IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. |
LINTING XUE et. al. | arxiv-cs.CL | 2020-10-22 |
| 402 | Sentence Boundary Augmentation For Neural Machine Translation Robustness IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Through in-depth error analysis, we show that sentence boundary segmentation has the largest impact on quality, and we develop a simple data augmentation strategy to improve segmentation robustness. |
Daniel Li; Te I; Naveen Arivazhagan; Colin Cherry; Dirk Padfield; | arxiv-cs.CL | 2020-10-21 |
| 403 | Beyond English-Centric Multilingual Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we create a true Many-to-Many multilingual translation model that can translate directly between any pair of 100 languages. We build and open source a training dataset that covers thousands of language directions with supervised data, created through large-scale mining. |
ANGELA FAN et. al. | arxiv-cs.CL | 2020-10-21 |
| 404 | A General Multi-Task Learning Framework to Leverage Text Data for Speech to Text Tasks IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this study, we propose a general multi-task learning framework to leverage text data for ASR and ST tasks. |
Yun Tang; Juan Pino; Changhan Wang; Xutai Ma; Dmitriy Genzel; | arxiv-cs.CL | 2020-10-21 |
| 405 | Cascaded Models With Cyclic Feedback For Direct Speech Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a technique that allows cascades of automatic speech recognition (ASR) and machine translation (MT) to exploit in-domain direct speech translation data in addition to out-of-domain MT and ASR data. |
Tsz Kin Lam; Shigehiko Schamoni; Stefan Riezler; | arxiv-cs.CL | 2020-10-21 |
| 406 | Analyzing The Source And Target Contributions To Predictions In Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We extend LRP to the Transformer and conduct an analysis of NMT models which explicitly evaluates the source and target relative contributions to the generation process. |
Elena Voita; Rico Sennrich; Ivan Titov; | arxiv-cs.CL | 2020-10-21 |
| 407 | On The Potential Of Lexico-logical Alignments For Semantic Parsing To SQL Queries IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose and test two methods: (1) supervised attention; (2) adopting an auxiliary objective of disambiguating references in the input queries to table columns. |
Tianze Shi; Chen Zhao; Jordan Boyd-Graber; Hal Daumé III; Lillian Lee; | arxiv-cs.CL | 2020-10-21 |
| 408 | Token Drop Mechanism For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose Token Drop to improve generalization and avoid overfitting for the NMT model. |
Huaao Zhang; Shigui Qiu; Xiangyu Duan; Min Zhang; | arxiv-cs.CL | 2020-10-21 |
| 409 | WaveTransformer: A Novel Architecture For Audio Captioning Based On Learning Temporal And Time-Frequency Information IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work we present a novel AAC novel method, explicitly focused on the exploitation of the temporal and time-frequency patterns in audio. |
An Tran; Konstantinos Drossos; Tuomas Virtanen; | arxiv-cs.SD | 2020-10-21 |
| 410 | TMT: A Transformer-based Modal Translator For Improving Multimodal Sequence Representations In Audio Visual Scene-aware Dialog IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Inspired by Neural Machine Translation (NMT), we propose the Transformer-based Modal Translator (TMT) to learn the representations of the source modal sequence by translating the source modal sequence to the related target modal sequence in a supervised manner. |
Wubo Li; Dongwei Jiang; Wei Zou; Xiangang Li; | arxiv-cs.CL | 2020-10-21 |
| 411 | Human-Paraphrased References Improve Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we compare the results of performing end-to-end system development using standard and paraphrased references. |
Markus Freitag; George Foster; David Grangier; Colin Cherry; | arxiv-cs.CL | 2020-10-20 |
| 412 | Complete Multilingual Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Multilingual Neural Machine Translation (MNMT) models are commonly trained on a joint set of bilingual corpora which is acutely English-centric (i.e. English either as the source or target language). |
Markus Freitag; Orhan Firat; | arxiv-cs.CL | 2020-10-20 |
| 413 | Word Shape Matters: Robust Machine Translation With Visual Embedding IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Neural machine translation has achieved remarkable empirical performance over standard benchmark datasets, yet recent evidence suggests that the models can still fail easily dealing with substandard inputs such as misspelled words, To overcome this issue, we introduce a new encoding heuristic of the input symbols for character-level NLP models: it encodes the shape of each character through the images depicting the letters when printed. |
Haohan Wang; Peiyan Zhang; Eric P. Xing; | arxiv-cs.CL | 2020-10-20 |
| 414 | Towards End-to-End In-Image Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we offer a preliminary investigation into the task of in-image machine translation: transforming an image containing text in one language into an image containing the same text in another language. |
ELMAN MANSIMOV et. al. | arxiv-cs.CL | 2020-10-20 |
| 415 | Pushing The Limits Of AMR Parsing With Self-Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore different ways in which trained models can be applied to improve AMR parsing performance, including generation of synthetic text and AMR annotations as well as refinement of actions oracle. |
YOUNG-SUK LEE et. al. | arxiv-cs.CL | 2020-10-20 |
| 416 | Unsupervised Pretraining For Neural Machine Translation Using Elastic Weight Consolidation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This work presents our ongoing research of unsupervised pretraining in neural machine translation (NMT). |
Dušan Variš; Ondřej Bojar; | arxiv-cs.CL | 2020-10-19 |
| 417 | Incorporating Terminology Constraints In Automatic Post-Editing IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Abstract: Users of machine translation (MT) may want to ensure the use of specific lexical terminologies. While there exist techniques for incorporating terminology constraints during … |
David Wan; Chris Kedzie; Faisal Ladhak; Marine Carpuat; Kathleen McKeown; | arxiv-cs.CL | 2020-10-19 |
| 418 | Diving Deep Into Context-Aware Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper analyzes the performance of document-level NMT models on four diverse domains with a varied amount of parallel document-level bilingual data. |
JINGJING HUO et. al. | arxiv-cs.CL | 2020-10-19 |
| 419 | Capturing Longer Context For Document-level Neural Machine Translation: A Multi-resolutional Approach IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose multi-resolutional (MR) Doc2Doc, a method to train a neural sequence-to-sequence model for document-level translation. |
ZEWEI SUN et. al. | arxiv-cs.CL | 2020-10-18 |
| 420 | Unsupervised Neural Machine Translation for Low-Resource Domains Via Meta-Learning IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address this issue, this paper presents a novel meta-learning algorithm for unsupervised neural machine translation (UNMT) that trains the model to adapt to another domain by utilizing only a small amount of training data. |
CHEONBOK PARK et. al. | arxiv-cs.CL | 2020-10-18 |
| 421 | Incorporate Semantic Structures Into Machine Translation Evaluation Via UCCA IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, in this work, we define words carrying important semantic meanings in sentences as semantic core words. |
Jin Xu; Yinuo Guo; Junfeng Hu; | arxiv-cs.CL | 2020-10-17 |
| 422 | A Corpus For English-Japanese Multimodal Neural Machine Translation With Comparable Sentences IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we propose a new multimodal English-Japanese corpus with comparable sentences that are compiled from existing image captioning datasets. |
Andrew Merritt; Chenhui Chu; Yuki Arase; | arxiv-cs.CL | 2020-10-17 |
| 423 | DiDi’s Machine Translation System For WMT2020 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes DiDi AI Labs’ submission to the WMT2020 news translation shared task. |
TANFANG CHEN et. al. | arxiv-cs.CL | 2020-10-16 |
| 424 | Unsupervised Bitext Mining And Translation Via Self-trained Contextual Embeddings IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We describe an unsupervised method to create pseudo-parallel corpora for machine translation (MT) from unaligned text. |
Phillip Keung; Julian Salazar; Yichao Lu; Noah A. Smith; | arxiv-cs.CL | 2020-10-15 |
| 425 | Multi-Task Learning For Cross-Lingual Abstractive Summarization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present a multi-task learning framework for cross-lingual abstractive summarization to augment training data. Recent studies constructed pseudo cross-lingual abstractive summarization data to train their neural encoder-decoders. Meanwhile, we introduce existing genuine data such as translation pairs and monolingual abstractive summarization data into training. |
Sho Takase; Naoaki Okazaki; | arxiv-cs.CL | 2020-10-15 |
| 426 | Mitigating Gender Bias In Machine Translation With Target Gender Annotations IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, in this work, we propose to decouple the task of acquiring the necessary information from the task of learning to translate correctly when such information is available. |
Artūrs Stafanovičs; Toms Bergmanis; Mārcis Pinnis; | arxiv-cs.CL | 2020-10-13 |
| 427 | Fine-grained Linguistic Evaluation For State-of-the-art Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes a test suite submission providing detailed statistics of linguistic performance for the state-of-the-art German-English systems of the Fifth Conference of Machine Translation (WMT20). |
Eleftherios Avramidis; Vivien Macketanz; Ursula Strohriegel; Aljoscha Burchardt; Sebastian Möller; | arxiv-cs.CL | 2020-10-13 |
| 428 | The Tatoeba Translation Challenge — Realistic Data Sets For Low Resource And Multilingual MT Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes the development of a new benchmark for machine translation that provides training and test data for thousands of language pairs covering over 500 languages and tools for creating state-of-the-art translation models from that collection. |
Jörg Tiedemann; | arxiv-cs.CL | 2020-10-13 |
| 429 | Improving Text Generation With Student-Forcing Optimal Transport IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To reduce this gap between training and testing, we propose using optimal transport (OT) to match the sequences generated in these two modes. |
GUOYIN WANG et. al. | arxiv-cs.CL | 2020-10-12 |
| 430 | Look It Up: Bilingual And Monolingual Dictionaries Improve Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we describe a new method for attaching dictionary definitions to rare words so that the network can learn the best way to use them. |
Xing Jie Zhong; David Chiang; | arxiv-cs.CL | 2020-10-12 |
| 431 | Gender Coreference And Bias Evaluation At WMT 2020 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Our work presents the largest evidence for the phenomenon in more than 19 systems submitted to the WMT over four diverse target languages: Czech, German, Polish, and Russian. |
Tom Kocmi; Tomasz Limisiewicz; Gabriel Stanovsky; | arxiv-cs.CL | 2020-10-12 |
| 432 | Collective Wisdom: Improving Low-resource Neural Machine Translation Using Adaptive Knowledge Distillation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we tackle this problem using knowledge distillation, where we propose to distill the knowledge of ensemble of teacher models to a single student model. |
Fahimeh Saleh; Wray Buntine; Gholamreza Haffari; | arxiv-cs.CL | 2020-10-12 |
| 433 | Retrieval Guided Unsupervised Multi-domain Image To Image Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we propose the use of an image retrieval system to assist the image-to-image translation task. |
RAUL GOMEZ et. al. | mm | 2020-10-12 |
| 434 | Towards Machine Translation For The Kurdish Language IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, in this paper, we are addressing the main issues in creating a machine translation system for the Kurdish language, with a focus on the Sorani dialect. |
Sina Ahmadi; Mariam Masoud; | arxiv-cs.CL | 2020-10-12 |
| 435 | Controllable Paraphrasing And Translation With A Syntactic Exemplar IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Despite only using bitext for training, and in near zero-shot conditions, our single proposed model can perform four tasks: controlled paraphrase generation in both languages and controlled machine translation in both language directions. To evaluate these tasks quantitatively, we create three novel evaluation datasets. |
Mingda Chen; Sam Wiseman; Kevin Gimpel; | arxiv-cs.CL | 2020-10-12 |
| 436 | Dynamic Context-guided Capsule Network For Multimodal Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To address the above issues, in this paper, we propose a novel Dynamic Context-guided Capsule Network (DCCN) for MMT. |
HUAN LIN et. al. | mm | 2020-10-12 |
| 437 | Fairseq S2T: Fast Speech-to-Text Modeling With Fairseq IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce fairseq S2T, a fairseq extension for speech-to-text (S2T) modeling tasks such as end-to-end speech recognition and speech-to-text translation. |
CHANGHAN WANG et. al. | arxiv-cs.CL | 2020-10-11 |
| 438 | Lexically Cohesive Neural Machine Translation With Copy Mechanism IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We employ a copy mechanism into a context-aware neural machine translation model to allow copying words from previous translation outputs. |
Vipul Mishra; Chenhui Chu; Yuki Arase; | arxiv-cs.CL | 2020-10-11 |
| 439 | Detecting Foodborne Illness Complaints In Multiple Languages Using English Annotations Only IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We demonstrate the benefits of our approach through extensive experiments with Yelp restaurant reviews in seven languages. |
Ziyi Liu; Giannis Karamanolakis; Daniel Hsu; Luis Gravano; | arxiv-cs.CL | 2020-10-11 |
| 440 | Addressing Exposure Bias With Document Minimum Risk Training: Cambridge At The WMT20 Biomedical Translation Task IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This can result in poor behaviour during inference if the model learns to neglect the source sentence. |
Danielle Saunders; Bill Byrne; | arxiv-cs.CL | 2020-10-11 |
| 441 | Neural Machine Translation Doesn’t Translate Gender Coreference Right Unless You Make It IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper we propose schemes for incorporating explicit word-level gender inflection tags into NMT. |
Danielle Saunders; Rosie Sallis; Bill Byrne; | arxiv-cs.CL | 2020-10-11 |
| 442 | Machine Translation Of Mathematical Text IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We have trained a Transformer-based translator model, using OpenNMT, on a combined corpus containing a small proportion of domain-specific sentences. |
Aditya Ohri; Tanya Schmah; | arxiv-cs.CL | 2020-10-11 |
| 443 | Automated Prediction Of Medieval Arabic Diacritics IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This study uses a character level neural machine translation approach trained on a long short-term memory-based bi-directional recurrent neural network architecture for diacritization of Medieval Arabic. |
Khalid Alnajjar; Mika Hämäläinen; Niko Partanen; Jack Rueter; | arxiv-cs.CL | 2020-10-11 |
| 444 | Localizing Open-Ontology QA Semantic Parsers In A Day Using Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose Semantic Parser Localizer (SPL), a toolkit that leverages Neural Machine Translation (NMT) systems to localize a semantic parser for a new language. |
Mehrad Moradshahi. Giovanni Campagna; Sina J. Semnani; Silei Xu; Monica S. Lam; | arxiv-cs.CL | 2020-10-10 |
| 445 | SJTU-NICT’s Supervised And Unsupervised Neural Machine Translation Systems For The WMT20 News Translation Task IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we introduced our joint team SJTU-NICT ‘s participation in the WMT 2020 machine translation shared task. |
ZUCHAO LI et. al. | arxiv-cs.CL | 2020-10-10 |
| 446 | Mark-Evaluate: Assessing Language Generation Using Population Estimation Methods IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose a family of metrics to assess language generation derived from population estimation methods widely used in ecology. |
Gonçalo Mordido; Christoph Meinel; | arxiv-cs.CL | 2020-10-09 |
| 447 | Recursive Top-Down Production For Sentence Generation With Latent Trees IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: To this end, we present a dynamic programming algorithm that marginalises over latent binary tree structures with $N$ leaves, allowing us to compute the likelihood of a sequence of $N$ tokens under a latent tree model, which we maximise to train a recursive neural function. |
Shawn Tan; Yikang Shen; Timothy J. O’Donnell; Alessandro Sordoni; Aaron Courville; | arxiv-cs.CL | 2020-10-09 |
| 448 | Word Level Language Identification In English Telugu Code Mixed Data IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a study of various models – Nave Bayes Classifier, Random Forest Classifier, Conditional Random Field (CRF), and Hidden Markov Model (HMM) for Language Identification in English – Telugu Code Mixed Data. |
Sunil Gundapu; Radhika Mamidi; | arxiv-cs.CL | 2020-10-09 |
| 449 | MLQE-PE: A Multilingual Quality Estimation And Post-Editing Dataset IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present MLQE-PE, a new dataset for Machine Translation (MT) Quality Estimation (QE) and Automatic Post-Editing (APE). |
MARINA FOMICHEVA et. al. | arxiv-cs.CL | 2020-10-09 |
| 450 | Learning To Evaluate Translation Beyond English: BLEURT Submissions To The WMT Metrics 2020 Shared Task IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: This paper describes our contribution to the WMT 2020 Metrics Shared Task, the main benchmark for automatic evaluation of translation. |
THIBAULT SELLAM et. al. | arxiv-cs.CL | 2020-10-08 |
| 451 | Leveraging Discourse Rewards For Document-Level Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Therefore, in this paper we propose a training approach that explicitly optimizes two established discourse metrics, lexical cohesion (LC) and coherence (COH), by using a reinforcement learning objective. |
Inigo Jauregi Unanue; Nazanin Esmaili; Gholamreza Haffari; Massimo Piccardi; | arxiv-cs.CL | 2020-10-07 |
| 452 | PyMT5: Multi-mode Translation Of Natural Language And Python Code With Transformers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We present an analysis and modeling effort of a large-scale parallel corpus of 26 million Python methods and 7.7 million method-docstring pairs, demonstrating that for docstring and method generation, PyMT5 outperforms similarly-sized auto-regressive language models (GPT2) which were English pre-trained or randomly initialized. |
Colin B. Clement; Dawn Drain; Jonathan Timcheck; Alexey Svyatkovskiy; Neel Sundaresan; | arxiv-cs.LG | 2020-10-07 |
| 453 | Scene Graph Modification Based On Natural Language Commands IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we explore the novel problem of graph modification, where the systems need to learn how to update an existing scene graph given a new user’s command. |
XUANLI HE et. al. | arxiv-cs.CL | 2020-10-06 |
| 454 | Why Skip If You Can Combine: A Simple Knowledge Distillation Technique For Intermediate Layers IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we target low-resource settings and evaluate our translation engines for Portuguese–English, Turkish–English, and English–German directions. |
Yimeng Wu; Peyman Passban; Mehdi Rezagholizade; Qun Liu; | arxiv-cs.CL | 2020-10-06 |
| 455 | On The Sparsity Of Neural Machine Translation Models IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: Modern neural machine translation (NMT) models employ a large number of parameters, which leads to serious over-parameterization and typically causes the underutilization of computational resources. |
Yong Wang; Longyue Wang; Victor O. K. Li; Zhaopeng Tu; | arxiv-cs.CL | 2020-10-06 |
| 456 | If Beam Search Is The Answer, What Was The Question? IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We frame beam search as the exact solution to a different decoding objective in order to gain insights into why high probability under a model alone may not indicate adequacy. |
Clara Meister; Tim Vieira; Ryan Cotterell; | arxiv-cs.CL | 2020-10-06 |
| 457 | An Empirical Study Of Tokenization Strategies For Various Korean NLP Tasks IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we test several tokenization strategies in order to answer our primary research question, that is, What is the best tokenization strategy for Korean NLP tasks? |
Kyubyong Park; Joohong Lee; Seongbo Jang; Dawoon Jung; | arxiv-cs.CL | 2020-10-06 |
| 458 | Pre-training Multilingual Neural Machine Translation By Leveraging Alignment Information IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We propose mRASP, an approach to pre-train a universal multilingual neural machine translation model. |
ZEHUI LIN et. al. | arxiv-cs.CL | 2020-10-06 |
| 459 | Multi-task Learning For Multilingual Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we propose a multi-task learning (MTL) framework that jointly trains the model with the translation task on bitext data and two denoising tasks on the monolingual data. |
Yiren Wang; ChengXiang Zhai; Hany Hassan Awadalla; | arxiv-cs.CL | 2020-10-06 |
| 460 | WikiLingua: A New Benchmark Dataset For Cross-Lingual Abstractive Summarization IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We introduce WikiLingua, a large-scale, multilingual dataset for the evaluation of crosslingual abstractive summarization systems. |
Faisal Ladhak; Esin Durmus; Claire Cardie; Kathleen McKeown; | arxiv-cs.CL | 2020-10-06 |
| 461 | Data Rejuvenation: Exploiting Inactive Training Examples For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we explore to identify the inactive training examples which contribute less to the model performance, and show that the existence of inactive examples depends on the data distribution. |
WENXIANG JIAO et. al. | arxiv-cs.CL | 2020-10-06 |
| 462 | Converting The Point Of View Of Messages Spoken To Virtual Assistants IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We developed a rule-based model, which integrates a linear text classification model, part-of-speech tagging, and constituency parsing with rule-based transformation methods. We have publicly released our dataset, which is composed of 46,565 crowd-sourced samples. |
Isabelle G. Lee; Vera Zu; Sai Srujana Buddi; Dennis Liang; Jack G. M. Fitzgerald; | arxiv-cs.CL | 2020-10-06 |
| 463 | We Don’t Speak The Same Language: Interpreting Polarization Through Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we present a new methodology that offers a fresh perspective on interpreting polarization through the lens of machine translation. |
Ashiqur R. KhudaBukhsh; Rupak Sarkar; Mark S. Kamlet; Tom M. Mitchell; | arxiv-cs.CL | 2020-10-05 |
| 464 | Adversarial Grammatical Error Correction IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we present an adversarial learning approach to GEC, using the generator-discriminator framework. |
Vipul Raheja; Dimitrios Alikaniotis; | arxiv-cs.CL | 2020-10-05 |
| 465 | Efficient Inference For Neural Machine Translation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this work, we look for the optimal combination of known techniques to optimize inference speed without sacrificing translation quality. |
Yi-Te Hsu; Sarthak Garg; Yi-Hsiu Liao; Ilya Chatsviorkin; | arxiv-cs.CL | 2020-10-05 |
| 466 | Participatory Research For Low-resourced Machine Translation: A Case Study In African Languages IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we focus on the task of Machine Translation (MT), that plays a crucial role for information accessibility and communication worldwide. |
WILHELMINA NEKOTO et. al. | arxiv-cs.CL | 2020-10-05 |
| 467 | Leveraging Multilingual News Websites For Building A Kurdish Parallel Corpus IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In an attempt to create parallel corpora for the Kurdish language, in this paper, we describe our approach in retrieving potentially-alignable news articles from multi-language websites and manually align them across dialects and languages based on lexical similarity and transliteration of scripts. We present a corpus containing 12,327 translation pairs in the two major dialects of Kurdish, Sorani and Kurmanji. |
Sina Ahmadi; Hossein Hassani; Daban Q. Jaff; | arxiv-cs.CL | 2020-10-04 |
| 468 | Syntax Representation In Word Embeddings And Neural Networks — A Survey Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: We describe which pre-trained models and representations of language are best suited for transfer to syntactic tasks. |
Tomasz Limisiewicz; David Mareček; | arxiv-cs.CL | 2020-10-02 |
| 469 | WeChat Neural Machine Translation Systems For WMT20 IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In our experiments, we employ data selection, several synthetic data generation approaches (i.e., back-translation, knowledge distillation, and iterative in-domain knowledge transfer), advanced finetuning approaches and self-bleu based model ensemble. |
FANDONG MENG et. al. | arxiv-cs.CL | 2020-10-01 |
| 470 | Learning Hard Retrieval Cross Attention For Transformer IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we accelerate the inference of the scaled dot-product attention in another perspective. |
Hongfei Xu; Qiuhui Liu; | arxiv-cs.CL | 2020-09-30 |
| 471 | A Simple But Tough-to-Beat Data Augmentation Approach For Natural Language Understanding And Generation IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we introduce a set of simple yet effective data augmentation strategies dubbed cutoff, where part of the information within an input sentence is erased to yield its restricted views (during the fine-tuning stage). |
Dinghan Shen; Mingzhi Zheng; Yelong Shen; Yanru Qu; Weizhu Chen; | arxiv-cs.CL | 2020-09-29 |
| 472 | Generative Latent Neural Models For Automatic Word Alignment IF:2 Related Papers Related Patents Related Grants Related Orgs Related Experts Details Highlight: In this paper, we study these models for the task of word alignment and propose and assess several evolutions of a vanilla variational autoencoders. |