Zobrazeno 1 - 10
of 22
pro vyhledávání: '"Roman Grundkiewicz"'
Publikováno v:
COLING (Tutorial)
Grammatical Error Correction (GEC) is the task of automatically detecting and correcting all types of errors in written text. Although most research has focused on correcting errors in the context of English as a Second Language (ESL), GEC can also b
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9c598a3e5eab365927073cf4645acd4c
https://doi.org/10.18653/v1/2020.coling-tutorials.6
https://doi.org/10.18653/v1/2020.coling-tutorials.6
Autor:
Marcin Junczys-Dowmunt, Chris Hokamp, Fabio Natanael Kepler, André F. T. Martins, Roman Grundkiewicz, Ramón Fernandez Astudillo
Publikováno v:
Martins, A F T, Junczys-Dowmunt, M, Kepler, F N, Astudillo, R, Hokamp, C & Grundkiewicz, R 2017, ' Pushing the Limits of Translation Quality Estimation ', vol. 5, pp. 205-218 .
Translation quality estimation is a task of growing importance in NLP, due to its potential to reduce post-editing human effort in disruptive ways. However, this potential is currently limited by the relatively low accuracy of existing systems. In th
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::adc11cf6795a4d8b94c9f8822153000c
https://doi.org/10.1162/tacl_a_00056
https://doi.org/10.1162/tacl_a_00056
Publikováno v:
Grundkiewicz, R & Junczys-Dowmuntz, M 2019, Minimally-Augmented Grammatical Error Correction . in Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019) . pp. 357–363, The 5th Workshop on Noisy User-generated Text (W-NUT): at EMNLP 2019, Hong Kong, 4/11/19 . https://doi.org/10.18653/v1/D19-5546
W-NUT@EMNLP
W-NUT@EMNLP
There has been an increased interest in low-resource approaches to automatic grammatical error correction. We introduce Minimally-Augmented Grammatical Error Correction (MAGEC) that does not require any error-labelled data. Our unsupervised approach
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ec06b425bad07b999aca11080fce1058
https://hdl.handle.net/20.500.11820/eea4e69d-9333-484c-aebf-996eb16b362b
https://hdl.handle.net/20.500.11820/eea4e69d-9333-484c-aebf-996eb16b362b
Autor:
Alham Fikri Aji, Young Jin Kim, Roman Grundkiewicz, Kenneth Heafield, Marcin Junczys-Dowmunt, Hany Hassan, Nikolay Bogoychev
Publikováno v:
NGT@EMNLP-IJCNLP
Proceedings of the 3rd Workshop on Neural Generation and Translation
Kim, Y J, Junczys-Dowmunt, M, Hassan, H, Aji, A F, Heafield, K, Grundkiewicz, R & Bogoychev, N 2019, From Research to Production and Back: Ludicrously Fast Neural Machine Translation . in Proceedings of the The 3rd Workshop on Neural Generation and Translation (WNGT 2019) . Hong Kong, pp. 280–288, The 3rd Workshop on Neural Generation and Translation, Hong Kong, Hong Kong, 4/11/19 . https://doi.org/10.18653/v1/D19-5632
Proceedings of the 3rd Workshop on Neural Generation and Translation
Kim, Y J, Junczys-Dowmunt, M, Hassan, H, Aji, A F, Heafield, K, Grundkiewicz, R & Bogoychev, N 2019, From Research to Production and Back: Ludicrously Fast Neural Machine Translation . in Proceedings of the The 3rd Workshop on Neural Generation and Translation (WNGT 2019) . Hong Kong, pp. 280–288, The 3rd Workshop on Neural Generation and Translation, Hong Kong, Hong Kong, 4/11/19 . https://doi.org/10.18653/v1/D19-5632
This paper describes the submissions of the “Marian” team to the WNGT 2019 efficiency shared task. Taking our dominating submissions to the previous edition of the shared task as a starting point, we develop improved teacher-student training via
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4b5768709a92eb40a46fed83d723003a
https://doi.org/10.18653/v1/d19-5632
https://doi.org/10.18653/v1/d19-5632
Publikováno v:
BEA@ACL
Grundkiewicz, R, Junczys-Dowmuntz, M & Heafield, K 2019, Neural Grammatical Error Correction Systems with Unsupervised Pre-training on Synthetic Data . in H Yannakoudakis, E Kochmar, C Leacock, N Madnani, I Pilán & T Zesch (eds), Proceedings of the Fourteenth Workshop on Innovative Use of NLP for Building Educational Applications ., W19-4427, Florence, Italy, pp. 252–263, 14th Workshop on Innovative Use of NLP for Building Educational Applications, Florence, Italy, 2/08/19 . < https://www.aclweb.org/anthology/W19-4427 >
University of Edinburgh-PURE
Grundkiewicz, R, Junczys-Dowmuntz, M & Heafield, K 2019, Neural Grammatical Error Correction Systems with Unsupervised Pre-training on Synthetic Data . in H Yannakoudakis, E Kochmar, C Leacock, N Madnani, I Pilán & T Zesch (eds), Proceedings of the Fourteenth Workshop on Innovative Use of NLP for Building Educational Applications ., W19-4427, Florence, Italy, pp. 252–263, 14th Workshop on Innovative Use of NLP for Building Educational Applications, Florence, Italy, 2/08/19 . < https://www.aclweb.org/anthology/W19-4427 >
University of Edinburgh-PURE
Considerable effort has been made to address the data sparsity problem in neural grammatical error correction. In this work, we propose a simple and surprisingly effective unsupervised synthetic error generation method based onconfusion sets extracte
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cfe94fc34cc550db34e627b7c34e5d58
https://doi.org/10.18653/v1/w19-4427
https://doi.org/10.18653/v1/w19-4427
Publikováno v:
University of Edinburgh-PURE
Junczys-Dowmunt, M, Heafield, K, Hoang, H, Grundkiewicz, R & Aue, A 2018, Marian: Cost-effective High-Quality Neural Machine Translation in C++ . in Proceedings of the 2nd Workshop on Neural Machine Translation and Generation . pp. 129-135, 2nd Workshop on Neural Machine Translation and Generation, Melbourne, Victoria, Australia, 15/07/18 . < http://aclweb.org/anthology/W18-2716 >
NMT@ACL
Junczys-Dowmunt, M, Heafield, K, Hoang, H, Grundkiewicz, R & Aue, A 2018, Marian: Cost-effective High-Quality Neural Machine Translation in C++ . in Proceedings of the 2nd Workshop on Neural Machine Translation and Generation . pp. 129-135, 2nd Workshop on Neural Machine Translation and Generation, Melbourne, Victoria, Australia, 15/07/18 . < http://aclweb.org/anthology/W18-2716 >
NMT@ACL
This paper describes the submissions of the "Marian" team to the WNMT 2018 shared task. We investigate combinations of teacher-student training, low-precision matrix products, auto-tuning and other methods to optimize the Transformer model on GPU and
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::529790a7a8ea3ddf8bfb2535a2fbe1b5
http://arxiv.org/abs/1805.12096
http://arxiv.org/abs/1805.12096
Autor:
Ulrich Germann, Alham Fikri Aji, Marcin Junczys-Dowmunt, Tom Neckermann, Frank Seide, Kenneth Heafield, Tomasz Dwojak, Alexandra Birch, Roman Grundkiewicz, Nikolay Bogoychev, Hieu Hoang, André F. T. Martins
Publikováno v:
arXiv.org e-Print Archive
Junczys-Dowmunt, M, Grundkiewicz, R, Dwojak, T, Hoang, H, Heafield, K, Neckermann, T, Seide, F, Germann, U, Aji, A F, Bogoychev, N, Martins, A F T & Birch, A 2018, Marian: Fast Neural Machine Translation in C++ . in Proceedings of ACL 2018, System Demonstrations . pp. 116–121, 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Australia, 15/07/18 . https://doi.org/10.18653/v1/P18-4020
ACL (4)
Junczys-Dowmunt, M, Grundkiewicz, R, Dwojak, T, Hoang, H, Heafield, K, Neckermann, T, Seide, F, Germann, U, Aji, A F, Bogoychev, N, Martins, A F T & Birch, A 2018, Marian: Fast Neural Machine Translation in C++ . in Proceedings of ACL 2018, System Demonstrations . pp. 116–121, 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Australia, 15/07/18 . https://doi.org/10.18653/v1/P18-4020
ACL (4)
We present Marian, an efficient and self-contained Neural Machine Translation framework with an integrated automatic differentiation engine based on dynamic computation graphs. Marian is written entirely in C++. We describe the design of the encoder-
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b8a4e0feaecdc54c56b8d2572462ec31
http://arxiv.org/abs/1804.00344
http://arxiv.org/abs/1804.00344
Publikováno v:
NAACL-HLT (2)
Grundkiewicz, R & Junczys-Dowmunt, M 2018, Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation . in Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) . pp. 284-290, 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, New Orleans, United States, 1/06/18 . https://doi.org/10.18653/v1/N18-2046
Grundkiewicz, R & Junczys-Dowmunt, M 2018, Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation . in Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) . pp. 284-290, 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, New Orleans, United States, 1/06/18 . https://doi.org/10.18653/v1/N18-2046
We combine two of the most popular approaches to automated Grammatical Error Correction (GEC): GEC based on Statistical Machine Translation (SMT) and GEC based on Neural Machine Translation (NMT). The hybrid system achieves new state-of-the-art resul
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c94f0be558a78f44fa52c48c8ee704ff
https://doi.org/10.18653/v1/n18-2046
https://doi.org/10.18653/v1/n18-2046
Publikováno v:
WMT (shared task)
This paper describes the Microsoft and University of Edinburgh submission to the Automatic Post-editing shared task at WMT2018. Based on training data and systems from the WMT2017 shared task, we re-implement our own models from the last shared task
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cf16de7f19301c68bc89bb1128474a03
Autor:
Kenneth Heafield, Roman Grundkiewicz
Publikováno v:
Grundkiewicz, R & Heafield, K 2018, Neural Machine Translation Techniques for Named Entity Transliteration . in Proceedings of the Seventh Named Entities Workshop . pp. 89–94, NEWS 2018: The Seventh Named Entities Workshop, Melbourne, Australia, 20/07/18 . < http://aclweb.org/anthology/W18-2413 >
University of Edinburgh-PURE
NEWS@ACL
University of Edinburgh-PURE
NEWS@ACL
Transliterating named entities from one language into another can be approached as neural machine translation (NMT) problem, for which we use deep attentional RNN encoder-decoder models. To build a strong transliteration system, we apply well-establi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3ddc7d21f52e15be15cec9519e3fe7f4
https://hdl.handle.net/20.500.11820/1b070b5e-f95f-4136-b065-195a41ad07dd
https://hdl.handle.net/20.500.11820/1b070b5e-f95f-4136-b065-195a41ad07dd