arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.07352cs.CL
收藏

翻译鲁棒性的视觉提示和纠错

Visual Cues and Error Correction for Translation Robustness

Imperial College London(帝国理工学院)

Zhenhao Li, Marek Rei, Lucia Specia

摘要:

神经机器翻译模型对输入文本中的噪声敏感,例如拼写错误的单词和不合语法的结构。现有的鲁棒性技术在面对不可见的噪声类型时通常会失败,并且它们的性能在干净的文本上会下降。在本文中,我们专注于三种类型的现实噪音,通常由人类产生的,并引入视觉上下文的想法,以提高翻译的鲁棒性噪声文本。此外,我们描述了一种新的纠错训练机制,可以作为一个辅助任务,以进一步提高翻译的鲁棒性。英法和英德翻译实验表明,多模态和纠错组件提高了模型对噪声文本的鲁棒性,同时仍然保留了干净文本的翻译质量。

英文摘要:

Neural Machine Translation models are sensitive to noise in the input texts, such as misspelled words and ungrammatical constructions. Existing robustness techniques generally fail when faced with unseen types of noise and their performance degrades on clean texts. In this paper, we focus on three types of realistic noise that are commonly generated by humans and introduce the idea of visual context to improve translation robustness for noisy texts. In addition, we describe a novel error correction training regime that can be used as an auxiliary task to further improve translation robustness. Experiments on English-French and English-German translation show that both multimodal and error correction components improve model robustness to noisy texts, while still retaining translation quality on clean texts.

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。

↑