Mention Attention for Pronoun Translation

TL;DR

This paper introduces a mention attention module in neural machine translation to improve pronoun translation by focusing on source mentions and incorporating target context, leading to better performance on pronoun translation tasks.

Contribution

The paper proposes a novel mention attention mechanism and mention classifiers to enhance pronoun translation in neural machine translation models.

Findings

Outperforms baseline Transformer in BLEU and APT scores

Improves pronoun translation accuracy without harming overall translation quality

Demonstrates effectiveness on WMT17 English-German translation task

Abstract

Most pronouns are referring expressions, computers need to resolve what do the pronouns refer to, and there are divergences on pronoun usage across languages. Thus, dealing with these divergences and translating pronouns is a challenge in machine translation. Mentions are referring candidates of pronouns and have closer relations with pronouns compared to general tokens. We assume that extracting additional mention features can help pronoun translation. Therefore, we introduce an additional mention attention module in the decoder to pay extra attention to source mentions but not non-mention tokens. Our mention attention module not only extracts features from source mentions, but also considers target-side context which benefits pronoun translation. In addition, we also introduce two mention classifiers to train models to recognize mentions, whose outputs guide the mention attention. We conduct experiments on the WMT17 English-German translation task, and evaluate our models on general translation and pronoun translation, using BLEU, APT, and contrastive evaluation metrics. Our proposed model outperforms the baseline Transformer model in terms of APT and BLEU scores, this confirms our hypothesis that we can improve pronoun translation by paying additional attention to source mentions, and shows that our introduced additional modules do not have negative effect on the general translation quality.