Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B
© 2019 Association for Computational Linguistics In this paper we propose a novel neural approach for automatic decipherment of lost languages. To compensate for the lack of strong supervision signal, our model design is informed by patterns in language change documented in historical linguistics. T...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Association for Computational Linguistics (ACL)
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137421.2 |
| _version_ | 1811077346928623616 |
|---|---|
| author | Luo, Jiaming Cao, Yuan Barzilay, Regina |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Luo, Jiaming Cao, Yuan Barzilay, Regina |
| author_sort | Luo, Jiaming |
| collection | MIT |
| description | © 2019 Association for Computational Linguistics In this paper we propose a novel neural approach for automatic decipherment of lost languages. To compensate for the lack of strong supervision signal, our model design is informed by patterns in language change documented in historical linguistics. The model utilizes an expressive sequence-to-sequence model to capture character-level correspondences between cognates. To effectively train the model in an unsupervised manner, we innovate the training procedure by formalizing it as a minimum-cost flow problem. When applied to the decipherment of Ugaritic, we achieve a 5.5% absolute improvement over state-of-the-art results. We also report the first automatic results in deciphering Linear B, a syllabic language related to ancient Greek, where our model correctly translates 67.3% of cognates. |
| first_indexed | 2024-09-23T10:41:30Z |
| format | Article |
| id | mit-1721.1/137421.2 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T10:41:30Z |
| publishDate | 2021 |
| publisher | Association for Computational Linguistics (ACL) |
| record_format | dspace |
| spelling | mit-1721.1/137421.22021-12-20T19:17:42Z Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B Luo, Jiaming Cao, Yuan Barzilay, Regina Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © 2019 Association for Computational Linguistics In this paper we propose a novel neural approach for automatic decipherment of lost languages. To compensate for the lack of strong supervision signal, our model design is informed by patterns in language change documented in historical linguistics. The model utilizes an expressive sequence-to-sequence model to capture character-level correspondences between cognates. To effectively train the model in an unsupervised manner, we innovate the training procedure by formalizing it as a minimum-cost flow problem. When applied to the decipherment of Ugaritic, we achieve a 5.5% absolute improvement over state-of-the-art results. We also report the first automatic results in deciphering Linear B, a syllabic language related to ancient Greek, where our model correctly translates 67.3% of cognates. Intelligence Advanced Research Projects Activity (Contract FA8650-17-C-9116) 2021-12-20T19:17:41Z 2021-11-05T11:30:57Z 2021-12-20T19:17:41Z 2019-08 2020-12-01T16:42:17Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137421.2 Luo, Jiaming, Cao, Yuan and Barzilay, Regina. 2019. "Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B." ACL 2019 - 57th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. en 10.18653/V1/P19-1303 ACL 2019 - 57th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference Creative Commons Attribution 4.0 International license https://creativecommons.org/licenses/by/4.0/ application/octet-stream Association for Computational Linguistics (ACL) Association for Computational Linguistics |
| spellingShingle | Luo, Jiaming Cao, Yuan Barzilay, Regina Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B |
| title | Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B |
| title_full | Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B |
| title_fullStr | Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B |
| title_full_unstemmed | Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B |
| title_short | Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B |
| title_sort | neural decipherment via minimum cost flow from ugaritic to linear b |
| url | https://hdl.handle.net/1721.1/137421.2 |
| work_keys_str_mv | AT luojiaming neuraldeciphermentviaminimumcostflowfromugaritictolinearb AT caoyuan neuraldeciphermentviaminimumcostflowfromugaritictolinearb AT barzilayregina neuraldeciphermentviaminimumcostflowfromugaritictolinearb |