Filters








1,066 Hits in 7.3 sec

Language Tags Matter for Zero-Shot Neural Machine Translation [article]

Liwei Wu, Shanbo Cheng, Mingxuan Wang, Lei Li
2021 arXiv   pre-print
Multilingual Neural Machine Translation (MNMT) has aroused widespread interest due to its efficiency.  ...  Experimental results show that by ignoring the source language tag (SLT) and adding the target language tag (TLT) to the encoder, the zero-shot translations could achieve a +8 BLEU score difference over  ...  Consistency by agreement in zero-shot neural machine translation.  ... 
arXiv:2106.07930v1 fatcat:2q22ow26erdbzjmfliseuxuaea

Cross-lingual Transfer Learning for Multilingual Task Oriented Dialog

Sebastian Schuster, Sonal Gupta, Rushin Shah, Mike Lewis
2019 Proceedings of the 2019 Conference of the North  
Since data collection for machine learning models for this task is time-consuming, it is desirable to make use of existing data in a high-resource language to train models in low-resource languages.  ...  machine translation encoder as contextual word representations.  ...  Acknowledgements We thank the three anonymous reviewers for their thoughtful comments.  ... 
doi:10.18653/v1/n19-1380 dblp:conf/naacl/SchusterGSL19 fatcat:yjsz3ckdj5gwtdi32hbcqbanvq

Cross-Lingual Transfer Learning for Multilingual Task Oriented Dialog [article]

Sebastian Schuster, Sonal Gupta, Rushin Shah, Mike Lewis
2019 arXiv   pre-print
Since data collection for machine learning models for this task is time-consuming, it is desirable to make use of existing data in a high-resource language to train models in low-resource languages.  ...  machine translation encoder as contextual word representations.  ...  Acknowledgements We thank the three anonymous reviewers for their thoughtful comments.  ... 
arXiv:1810.13327v2 fatcat:6urxuwur45hylfaootgcnbfh5a

A Study of Multilingual Neural Machine Translation [article]

Xu Tan, Yichong Leng, Jiale Chen, Yi Ren, Tao Qin, Tie-Yan Liu
2019 arXiv   pre-print
Multilingual neural machine translation (NMT) has recently been investigated from different aspects (e.g., pivot translation, zero-shot translation, fine-tuning, or training from scratch) and in different  ...  smaller; (5) given a fixed training data budget, it is better to introduce more languages into multilingual training for zero-shot translation.  ...  It is well known that adding similar languages would benefit for the zero-shot translation.  ... 
arXiv:1912.11625v1 fatcat:ir4iqit27fdbrntqnbwc67fydu

A Comprehensive Survey of Multilingual Neural Machine Translation [article]

Raj Dabre, Chenhui Chu, Anoop Kunchukuttan
2020 arXiv   pre-print
We present a survey on multilingual neural machine translation (MNMT), which has gained a lot of traction in the recent years.  ...  MNMT is more promising and interesting than its statistical machine translation counterpart because end-to-end modeling and distributed representations open new avenues for research on machine translation  ...  Limitations of Zero-shot Translation.  ... 
arXiv:2001.01115v2 fatcat:rkl6utbsufdbpjvawx6vrqspum

Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsing [article]

Tal Schuster, Ori Ram, Regina Barzilay, Amir Globerson
2019 arXiv   pre-print
Our experimental results demonstrate the effectiveness of this approach for zero-shot and few-shot learning of dependency parsing.  ...  This mapping readily supports processing of a target language, improving transfer by context-aware embeddings.  ...  Acknowledgements We thank the MIT NLP group and the reviewers for their helpful discussion and comments. 11 We filtered words with multiple translations to the most common one by Google Translate.  ... 
arXiv:1902.09492v2 fatcat:hlhz4gbikfaejpucvcfhptoblq

Examining Scaling and Transfer of Language Model Architectures for Machine Translation [article]

Biao Zhang, Behrooz Ghorbani, Ankur Bapna, Yong Cheng, Xavier Garcia, Jonathan Shen, Orhan Firat
2022 arXiv   pre-print
bilingual and multilingual translation tasks, and improve greatly on zero-shot directions by facilitating the reduction of off-target translations.  ...  In this work, we thoroughly examine the role of several architectural design choices on the performance of LMs on bilingual, (massively) multilingual and zero-shot translation tasks, under systematic variations  ...  parameters, and designed different language tags for multilingual translation that could greatly affect zero-shot results (Wu et al., 2021) .  ... 
arXiv:2202.00528v3 fatcat:jlzm5kxssvamzmh2a3g43oknya

Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsing

Tal Schuster, Ori Ram, Regina Barzilay, Amir Globerson
2019 Proceedings of the 2019 Conference of the North  
Our experimental results demonstrate the effectiveness of this approach for zero-shot and few-shot learning of dependency parsing.  ...  This mapping readily supports processing of a target language, improving transfer by context-aware embeddings.  ...  Acknowledgements We thank the MIT NLP group and the reviewers for their helpful discussion and comments. 11 We filtered words with multiple translations to the most common one by Google Translate.  ... 
doi:10.18653/v1/n19-1162 dblp:conf/naacl/SchusterRBG19 fatcat:ns2bxzatkjdovnyqzxegtw53i4

Structure-Level Knowledge Distillation For Multilingual Sequence Labeling [article]

Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Fei Huang, Kewei Tu
2020 arXiv   pre-print
Multilingual sequence labeling is a task of predicting label sequences using a single unified model for multiple languages.  ...  Our experiments on 4 multilingual tasks with 25 datasets show that our approaches outperform several strong baselines and have stronger zero-shot generalizability than both the baseline model and teacher  ...  Pooled contextualized embeddings for named  ... 
arXiv:2004.03846v3 fatcat:od6g6j57gjfddb3mqy2phtiewy

Getting Gender Right in Neural Machine Translation

Eva Vanmassenhove, Christian Hardmeier, Andy Way
2018 Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing  
Our experiments show that adding a gender feature to an NMT system significantly improves the translation quality for some language pairs.  ...  Our contribution is two-fold: (1) the compilation of large datasets with speaker information for 20 language pairs, and (2) a simple set of experiments that incorporate gender information into NMT for  ...  Acknowledgements We would also like to thank the anonymous reviewers for their insightful comments and feedback.  ... 
doi:10.18653/v1/d18-1334 dblp:conf/emnlp/VanmassenhoveHW18 fatcat:k5zcmy3xivenbngxih3gnmmvpa

Zero-Shot Cross-Lingual Transfer with Meta Learning [article]

Farhad Nooralahzadeh, Giannis Bekoulis, Johannes Bjerva, Isabelle Augenstein
2020 arXiv   pre-print
We improve upon the state-of-the-art for zero-shot and few-shot NLI (on MultiNLI and XNLI) and QA (on the MLQA dataset).  ...  We experiment using standard supervised, zero-shot cross-lingual, as well as few-shot cross-lingual settings for different natural language understanding tasks (natural language inference, question answering  ...  We are grateful to the Nordic Language Processing Laboratory (NLPL) for providing access to its supercluster infrastructure.  ... 
arXiv:2003.02739v4 fatcat:zry6xjzgznabnah3xyx4btdjim

A Survey of Multilingual Neural Machine Translation

Raj Dabre, Chenhui Chu, Anoop Kunchukuttan
2020 ACM Computing Surveys  
We present a survey on multilingual neural machine translation (MNMT), which has gained a lot of traction in recent years.  ...  MNMT is more promising and interesting than its statistical machine translation counterpart, because end-to-end modeling and distributed representations open new avenues for research on machine translation  ...  ACKNOWLEDGMENTS We thank the anonymous reviewers for their insightful comments.  ... 
doi:10.1145/3406095 fatcat:5fha6uwpqjeubbwwoxcdd26bge

Extreme Classification (Dagstuhl Seminar 18291)

Samy Bengio, Krzysztof Dembczynski, Thorsten Joachims, Marius Kloft, Manik Varma, Michael Wagner
2019 Dagstuhl Reports  
Many applications of extreme classification have been found in diverse areas ranging from language modeling to document tagging in NLP, face recognition to learning universal feature representations in  ...  Extreme classification has also opened up a new paradigm for key industrial applications such as ranking and recommendation by reformulating them as multi-label learning tasks where each item to be ranked  ...  We discuss potential solutions for this problem, such as: using character level information, copy mechanisms (e.g. in machine translation) or few shot learning with cache models.  ... 
doi:10.4230/dagrep.8.7.62 dblp:journals/dagstuhl-reports/BengioDJKV18 fatcat:tglxen4d4vc5vkxtllzy3xokl4

"Wikily" Supervised Neural Translation Tailored to Cross-Lingual Tasks [article]

Mohammad Sadegh Rasooli, Chris Callison-Burch, Derry Tanti Wijaya
2021 arXiv   pre-print
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  ...  In image captioning, we train a multi-tasking machine translation and image captioning pipeline for Arabic and English from which the Arabic training data is a translated version of the English captioning  ...  Acknowledgments We would like to thank reviewers and the editor for their useful comments.  ... 
arXiv:2104.08384v2 fatcat:vswaqg27mve4fpepwuxqzougru

HintedBT: Augmenting Back-Translation with Quality and Transliteration Hints [article]

Sahana Ramnath, Melvin Johnson, Abhirut Gupta, Aravindan Raghuveer
2021 arXiv   pre-print
Back-translation (BT) of target monolingual corpora is a widely used data augmentation strategy for neural machine translation (NMT), especially for low-resource language pairs.  ...  For such cases, we propose training the model with additional hints (as target tags on the decoder) that provide information about the operation required on the source (translation or both translation  ...  We also thank the reviewers for their valuable and constructive suggestions.  ... 
arXiv:2109.04443v1 fatcat:ia65n4wbwbd3touhdw6754tv2u
« Previous Showing results 1 — 15 out of 1,066 results