Graph Algorithms for Multiparallel Word Alignment

TL;DR

This paper introduces graph-based algorithms for leveraging multiparallel corpora to improve word alignment accuracy in machine translation, demonstrating significant F1 score improvements over traditional methods.

Contribution

It presents novel graph algorithms inspired by recommender systems and network link prediction to exploit multiparallel data for enhanced word alignment.

Findings

Up to 28% absolute F1 improvement over baseline aligner

Effective utilization of multiparallel corpora for alignment

Demonstrated across multiple datasets

Abstract

With the advent of end-to-end deep learning approaches in machine translation, interest in word alignments initially decreased; however, they have again become a focus of research more recently. Alignments are useful for typological research, transferring formatting like markup to translated texts, and can be used in the decoding of machine translation systems. At the same time, massively multilingual processing is becoming an important NLP scenario, and pretrained language and machine translation models that are truly multilingual are proposed. However, most alignment algorithms rely on bitexts only and do not leverage the fact that many parallel corpora are multiparallel. In this work, we exploit the multiparallelity of corpora by representing an initial set of bilingual alignments as a graph and then predicting additional edges in the graph. We present two graph algorithms for edge prediction: one inspired by recommender systems and one based on network link prediction. Our experimental results show absolute improvements in $F_1$ of up to 28% over the baseline bilingual word aligner in different datasets.