Agreement-based Learning of Parallel Lexicons and Phrases from Non-Parallel Corpora

TL;DR

This paper presents an agreement-based method for learning parallel lexicons and phrases from non-parallel corpora by encouraging two translation models to agree on alignments, resulting in improved performance.

Contribution

It introduces a novel agreement-based learning framework with a Viterbi EM algorithm for joint training of unidirectional translation models from non-parallel data.

Findings

Significant improvement in alignment accuracy

Enhanced translation performance on Chinese-English data

Effective joint training of asymmetric models

Abstract

We introduce an agreement-based approach to learning parallel lexicons and phrases from non-parallel corpora. The basic idea is to encourage two asymmetric latent-variable translation models (i.e., source-to-target and target-to-source) to agree on identifying latent phrase and word alignments. The agreement is defined at both word and phrase levels. We develop a Viterbi EM algorithm for jointly training the two unidirectional models efficiently. Experiments on the Chinese-English dataset show that agreement-based learning significantly improves both alignment and translation performance.