Embedding-Enhanced Giza++: Improving Alignment in Low- and High- Resource Scenarios Using Embedding Space Geometry

Abstract

A popular natural language processing task decades ago, word alignment has been dominated until recently by GIZA++, a statistical method based on the 30-year-old IBM models. New methods that outperform GIZA++ primarily rely on large machine translation models, massively multilingual language models, or supervision from GIZA++ alignments itself. We introduce Embedding-Enhanced GIZA++, and outperform GIZA++ without any of the aforementioned factors. Taking advantage of monolingual embedding spaces of source and target language only, we exceed GIZA++'s performance in every tested scenario for three languages pairs. In the lowest-resource setting, we outperform GIZA++ by 8.5, 10.9, and 12 AER for Ro-En, De-En, and En-Fr, respectively. We release our code at https://github.com/kellymarchisio/ee-giza.

AMTA2022 Camera Ready

References in corpus (5)

• Distributed Representations of Words and Phrases and their Compositionality
• Adding Interpretable Attention to Neural Translation Models Improves Word Alignment
• Loss in Translation: Learning Bilingual Word Mapping with a Retrieval Criterion
• Word Alignment by Fine-tuning Embeddings on Parallel Corpora
• On a Novel Application of Wasserstein-Procrustes for Unsupervised Cross-Lingual Learning

Cited by in corpus (1)

• Graph Algorithms for Multiparallel Word Alignment