arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.08473cs.CL
收藏

多视角对齐提升神经机器翻译的自然度

Multi-perspective Alignment for Increasing Naturalness in Neural Machine Translation

CLCG, University of Groningen(格罗宁根大学CLCG);Department of Computer Science, Aalborg University(奥尔堡大学计算机科学系);Departament de Llenguatges i Sistemes Informàtics, Universitat d’Alacant(阿莱康特大学语言与信息系统系)

Huiyuan Lai, Esther Ploeger, Rik van Noord, Antonio Toral

AI总结:

受人类反馈强化学习框架启发,提出一种多视角对齐方法,在英译荷兰语文学翻译中同时奖励自然度和内容保留,生成词汇更丰富且不损失准确性的翻译。

摘要:

神经机器翻译(NMT)系统会放大其训练数据中存在的词汇偏差,导致输出翻译中的语言被人为地贫乏化。这些语言层面的特征使得自动翻译不同于原本用某种语言书写的文本和人类翻译,这阻碍了它们在例如创建评估数据集等方面的有用性。增加NMT自然度的尝试在内容保留方面可能存在不足,其中增加的词汇多样性是以牺牲翻译准确性为代价的。受人类反馈强化学习框架的启发,我们引入了一种新方法,该方法同时对自然度和内容保留进行奖励。我们尝试从多个视角来产生更自然的翻译,旨在减少机器和人类翻译腔。我们在英译荷兰语文学翻译上评估了我们的方法,并发现我们的最佳模型产生的翻译在词汇上更丰富,并表现出更多人类书写语言的特性,且没有损失翻译准确性。

英文摘要:

Neural machine translation (NMT) systems amplify lexical biases present in their training data, leading to artificially impoverished language in output translations. These language-level characteristics render automatic translations different from text originally written in a language and human translations, which hinders their usefulness in for example creating evaluation datasets. Attempts to increase naturalness in NMT can fall short in terms of content preservation, where increased lexical diversity comes at the cost of translation accuracy. Inspired by the reinforcement learning from human feedback framework, we introduce a novel method that rewards both naturalness and content preservation. We experiment with multiple perspectives to produce more natural translations, aiming at reducing machine and human translationese. We evaluate our method on English-to-Dutch literary translation, and find that our best model produces translations that are lexically richer and exhibit more properties of human-written language, without loss in translation accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑