Using a vocabulary that is shared across languages is common practice in Multilingual Neural Machine Translation (MNMT). In addition to its simple design, shared tokens play an important role in positive knowledge transfer, assuming that shared tokens refer to similar meanings across languages. However, when word overlap is small, especially due to different writing systems, transfer is inhibited. In this paper, we define word-level information transfer pathways via word equivalence classes and rely on graph networks to fuse word embeddings across languages. Our experiments demonstrate the advantages of our approach: 1) embeddings of words with similar meanings are better aligned across languages, 2) our method achieves consistent BLEU improvements of up to 2.3 points for high- and low-resource MNMT, and 3) less than 1.0\% additional trainable parameters are required with a limited increase in computational costs, while inference time remains identical to the baseline. We release the codebase to the community.

翻译：在多语言神经机器翻译（Multilingual Neural Machine Translation, MNMT）中，使用跨语言共享词汇是一种常见实践。除设计简洁外，共享词元在正向知识迁移中发挥关键作用——其前提是共享词元在不同语言中指向相似含义。然而，当词汇重叠度较低（尤其因书写系统差异）时，知识迁移会受到抑制。本文通过词等价类定义词语级信息迁移路径，并借助图网络融合跨语言词嵌入。实验表明本方法具有以下优势：1）语义相近的词嵌入在跨语言间对齐更优；2）在高资源与低资源MNMT场景中，本方法可实现最高2.3个BLEU点的持续提升；3）新增可训练参数不足1.0%，计算开销增幅有限，且推理速度与基线持平。我们已将代码库开源。