Learning to Create and Reuse Words in Open-Vocabulary Neural Language Modeling

Kazuya Kawakami, Chris Dyer, Phil Blunsom


Abstract
Fixed-vocabulary language models fail to account for one of the most characteristic statistical facts of natural language: the frequent creation and reuse of new word types. Although character-level language models offer a partial solution in that they can create word types not attested in the training corpus, they do not capture the “bursty” distribution of such words. In this paper, we augment a hierarchical LSTM language model that generates sequences of word tokens character by character with a caching mechanism that learns to reuse previously generated words. To validate our model we construct a new open-vocabulary language modeling corpus (the Multilingual Wikipedia Corpus; MWC) from comparable Wikipedia articles in 7 typologically diverse languages and demonstrate the effectiveness of our model across this range of languages.
Anthology ID:
P17-1137
Volume:
Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2017
Address:
Vancouver, Canada
Editors:
Regina Barzilay, Min-Yen Kan
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1492–1502
Language:
URL:
https://aclanthology.org/P17-1137
DOI:
10.18653/v1/P17-1137
Bibkey:
Cite (ACL):
Kazuya Kawakami, Chris Dyer, and Phil Blunsom. 2017. Learning to Create and Reuse Words in Open-Vocabulary Neural Language Modeling. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1492–1502, Vancouver, Canada. Association for Computational Linguistics.
Cite (Informal):
Learning to Create and Reuse Words in Open-Vocabulary Neural Language Modeling (Kawakami et al., ACL 2017)
Copy Citation:
PDF:
https://aclanthology.org/P17-1137.pdf
Dataset:
 P17-1137.Datasets.zip
Data
WikiText-2