Searching for Search Errors in Neural Morphological Inflection

Martina Forster, Clara Meister, Ryan Cotterell


Abstract
Neural sequence-to-sequence models are currently the predominant choice for language generation tasks. Yet, on word-level tasks, exact inference of these models reveals the empty string is often the global optimum. Prior works have speculated this phenomenon is a result of the inadequacy of neural models for language generation. However, in the case of morphological inflection, we find that the empty string is almost never the most probable solution under the model. Further, greedy search often finds the global optimum. These observations suggest that the poor calibration of many neural models may stem from characteristics of a specific subset of tasks rather than general ill-suitedness of such models for language generation.
Anthology ID:
2021.eacl-main.118
Volume:
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Month:
April
Year:
2021
Address:
Online
Editors:
Paola Merlo, Jorg Tiedemann, Reut Tsarfaty
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1388–1394
Language:
URL:
https://aclanthology.org/2021.eacl-main.118
DOI:
10.18653/v1/2021.eacl-main.118
Bibkey:
Cite (ACL):
Martina Forster, Clara Meister, and Ryan Cotterell. 2021. Searching for Search Errors in Neural Morphological Inflection. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 1388–1394, Online. Association for Computational Linguistics.
Cite (Informal):
Searching for Search Errors in Neural Morphological Inflection (Forster et al., EACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.eacl-main.118.pdf