Neural machine translation (NMT) has achieved remarkable success in producing high-quality translations. However, current NMT systems suffer from a lack of reliability, as their outputs that are often affected by lexical or syntactic changes in inputs, resulting in large variations in quality. This limitation hinders the practicality and trustworthiness of NMT. A contributing factor to this problem is that NMT models trained with the one-to-one paradigm struggle to handle the source diversity phenomenon, where inputs with the same meaning can be expressed differently. In this work, we treat this problem as a bilevel optimization problem and present a consistency-aware meta-learning (CAML) framework derived from the model-agnostic meta-learning (MAML) algorithm to address it. Specifically, the NMT model with CAML (named CoNMT) first learns a consistent meta representation of semantically equivalent sentences in the outer loop. Subsequently, a mapping from the meta representation to the output sentence is learned in the inner loop, allowing the NMT model to translate semantically equivalent sentences to the same target sentence. We conduct experiments on the NIST Chinese to English task, three WMT translation tasks, and the TED M2O task. The results demonstrate that CoNMT effectively improves overall translation quality and reliably handles diverse inputs.

翻译：神经机器翻译（NMT）在生成高质量译文方面取得了显著成功。然而，当前NMT系统缺乏可靠性，其输出常受输入词汇或句法变化的影响，导致质量出现大幅波动。这一局限性阻碍了NMT的实用性和可信度。造成该问题的因素之一是，采用一对一句式训练范式所得的NMT模型难以应对源语言多样性现象——即具有相同语义的输入可能以不同形式表达。在本工作中，我们将该问题视为双层优化问题，并提出一种基于模型无关元学习（MAML）算法的一致性感知元学习（CAML）框架加以解决。具体而言，采用CAML的NMT模型（命名为CoNMT）首先在外循环中学习语义等价句子的统一元表示。随后，在内循环中学习从元表示到输出句子的映射，使NMT模型能够将语义等价的句子翻译为相同的目标译文。我们在NIST中英翻译任务、三项WMT翻译任务及TED M2O任务上开展实验。结果表明，CoNMT有效提升了整体翻译质量，并能够可靠地处理多样化输入。