Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings
In the absence of annotations in the target language, multilingual models typically draw on extensive parallel resources. In this paper, we demonstrate that accurate multilingual partof-speech (POS) tagging can be done with just a few (e.g., ten) word translation pairs. We use the translation pairs...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Association for Computational Linguistics
2017
|
| Online Access: | http://hdl.handle.net/1721.1/110739 https://orcid.org/0000-0003-3121-0185 https://orcid.org/0000-0002-2921-8201 https://orcid.org/0000-0002-2199-0379 |
| _version_ | 1826214964694614016 |
|---|---|
| author | Gaddy, David M. Zhang, Yuan Barzilay, Regina Jaakkola, Tommi S. |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Gaddy, David M. Zhang, Yuan Barzilay, Regina Jaakkola, Tommi S. |
| author_sort | Gaddy, David M. |
| collection | MIT |
| description | In the absence of annotations in the target language, multilingual models typically draw on extensive parallel resources. In this paper, we
demonstrate that accurate multilingual partof-speech (POS) tagging can be done with just a few (e.g., ten) word translation pairs. We use the translation pairs to establish a coarse linear isometric (orthonormal) mapping between
monolingual embeddings. This enables the supervised source model expressed in terms of embeddings to be used directly on the target language. We further refine the model in an unsupervised manner by initializing and regularizing it to be close to the direct transfer model. Averaged across six languages, our model yields a 37.5% absolute
improvement over the monolingual prototypedriven method (Haghighi and Klein, 2006) when using a comparable amount of supervision. Moreover, to highlight key linguistic characteristics of the generated tags, we use them to predict typological properties of languages, obtaining a 50% error reduction relative to the prototype model |
| first_indexed | 2024-09-23T16:14:25Z |
| format | Article |
| id | mit-1721.1/110739 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:14:25Z |
| publishDate | 2017 |
| publisher | Association for Computational Linguistics |
| record_format | dspace |
| spelling | mit-1721.1/1107392022-09-29T19:03:35Z Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings Gaddy, David M. Zhang, Yuan Barzilay, Regina Jaakkola, Tommi S. Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Gaddy, David M. Zhang, Yuan Barzilay, Regina Jaakkola, Tommi S. In the absence of annotations in the target language, multilingual models typically draw on extensive parallel resources. In this paper, we demonstrate that accurate multilingual partof-speech (POS) tagging can be done with just a few (e.g., ten) word translation pairs. We use the translation pairs to establish a coarse linear isometric (orthonormal) mapping between monolingual embeddings. This enables the supervised source model expressed in terms of embeddings to be used directly on the target language. We further refine the model in an unsupervised manner by initializing and regularizing it to be close to the direct transfer model. Averaged across six languages, our model yields a 37.5% absolute improvement over the monolingual prototypedriven method (Haghighi and Klein, 2006) when using a comparable amount of supervision. Moreover, to highlight key linguistic characteristics of the generated tags, we use them to predict typological properties of languages, obtaining a 50% error reduction relative to the prototype model 2017-07-17T18:13:57Z 2017-07-17T18:13:57Z 2016-06 Article http://purl.org/eprint/type/ConferencePaper 978-1-941643-91-4 http://hdl.handle.net/1721.1/110739 Zhang, Yuan et al. "Ten Pairs to Tag - Multilingual POS Tagging via Course Mapping between Embeddings." 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, San Diego, California, USA, 12-17 June, 2016. Association for Computational Linguistics, 2016. https://orcid.org/0000-0003-3121-0185 https://orcid.org/0000-0002-2921-8201 https://orcid.org/0000-0002-2199-0379 en_US http://dblp.dagstuhl.de/db/conf/naacl/naacl2016.html 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for Computational Linguistics MIT Web Domain |
| spellingShingle | Gaddy, David M. Zhang, Yuan Barzilay, Regina Jaakkola, Tommi S. Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings |
| title | Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings |
| title_full | Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings |
| title_fullStr | Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings |
| title_full_unstemmed | Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings |
| title_short | Ten pairs to tag - Multilingual POS tagging via coarse mapping between embeddings |
| title_sort | ten pairs to tag multilingual pos tagging via coarse mapping between embeddings |
| url | http://hdl.handle.net/1721.1/110739 https://orcid.org/0000-0003-3121-0185 https://orcid.org/0000-0002-2921-8201 https://orcid.org/0000-0002-2199-0379 |
| work_keys_str_mv | AT gaddydavidm tenpairstotagmultilingualpostaggingviacoarsemappingbetweenembeddings AT zhangyuan tenpairstotagmultilingualpostaggingviacoarsemappingbetweenembeddings AT barzilayregina tenpairstotagmultilingualpostaggingviacoarsemappingbetweenembeddings AT jaakkolatommis tenpairstotagmultilingualpostaggingviacoarsemappingbetweenembeddings |