Quality-Aware Decoding for Neural Machine Translation
Quality-Aware Decoding for Neural Machine Translation
复制标题
神经机器翻译的质量感知解码
DOI：
10.48550/arxiv.2205.00978
复制
发表时间：
2022
影响因子：
10.9
通讯作者：
André F. T. Martins
中科院分区：
文献类型：
--
作者：
Patrick Fernandes;António Farinhas;Ricardo Rei;José G. C. de Souza;Perez Ogayo;Graham Neubig;André F. T. Martins
尽管近年来在机器翻译质量评估和评估方面取得了进展，但神经机器翻译（NMT）中的解码大多忽略了这一点，并围绕着根据模型（MAP解码）找到最可能的翻译，近似于波束搜索。在本文中，我们将这两条研究路线结合在一起，并通过各种推理方法（如N-最佳重新排序和最小贝叶斯风险解码），利用最近在无参考和基于参考的MT评估方面的突破，提出了NMT的质量感知解码。我们在四个数据集和两个模型类中对各种可能的候选生成和排名方法进行了广泛的比较，发现质量感知解码始终优于基于MAP的解码，无论是根据最先进的自动度量（COMET和BLEURT）还是人工评估。
Despite the progress in machine translation quality estimation and evaluation in the last years, decoding in neural machine translation (NMT) is mostly oblivious to this and centers around finding the most probable translation according to the model (MAP decoding), approximated with beam search. In this paper, we bring together these two lines of research and propose quality-aware decoding for NMT, by leveraging recent breakthroughs in reference-free and reference-based MT evaluation through various inference methods like N-best reranking and minimum Bayes risk decoding. We perform an extensive comparison of various possible candidate generation and ranking methods across four datasets and two model classes and find that quality-aware decoding consistently outperforms MAP-based decoding according both to state-of-the-art automatic metrics (COMET and BLEURT) and to human assessments.