Robust Open-Vocabulary Translation from Visual Text Representations

Elizabeth Salesky, David Etter, Matt Post


Abstract
Machine translation models have discrete vocabularies and commonly use subword segmentation techniques to achieve an ‘open vocabulary.’ This approach relies on consistent and correct underlying unicode sequences, and makes models susceptible to degradation from common types of noise and variation. Motivated by the robustness of human language processing, we propose the use of visual text representations, which dispense with a finite set of text embeddings in favor of continuous vocabularies created by processing visually rendered text with sliding windows. We show that models using visual text representations approach or match performance of traditional text models on small and larger datasets. More importantly, models with visual embeddings demonstrate significant robustness to varied types of noise, achieving e.g., 25.9 BLEU on a character permuted German–English task where subword models degrade to 1.9.
Anthology ID:
2021.emnlp-main.576
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7235–7252
Language:
URL:
https://aclanthology.org/2021.emnlp-main.576
DOI:
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.576.pdf
Code
 esalesky/visrep
Data
MTNT