Scalable Inference and Training of Context-Rich Syntactic Translation Models
Scalable Inference and Training of Context-Rich Syntactic Translation Models
复制标题
DOI：
10.3115/1220175.1220296
复制
发表时间：
2006-07
期刊：
影响因子：
--
通讯作者：
Michel Galley;Jonathan Graehl;Kevin Knight;D. Marcu;Steve DeNeefe;Wei Wang;I. Thayer
中科院分区：
文献类型：
--
作者：
Michel Galley;Jonathan Graehl;Kevin Knight;D. Marcu;Steve DeNeefe;Wei Wang;I. Thayer
统计机器翻译在过去几年中取得了长足的进步，但当前的翻译模型在重新排序和目标语言流畅性方面较弱。句法方法试图解决这些问题。在本文中，我们采用从对齐的树串对获取多级句法翻译规则的框架（Galley et al., 2004），并提出了其方法的两个主要扩展：首先，我们构造了大量包含上下文更丰富的规则的推导，而不是仅仅计算最低限度解释句子对的单个推导，并考虑了未对齐单词的多种解释。其次，我们提出概率估计和对这些规则进行加权的训练程序。我们在实际例子中对比了不同的方法，表明我们基于多重推导的估计有利于在语言上更好地激发的短语重新排序，并确定我们的较大规则比最小规则提供了 3.63 BLEU 点的增加。
Statistical MT has made great progress in the last few years, but current translation models are weak on re-ordering and target language fluency. Syntactic approaches seek to remedy these problems. In this paper, we take the framework for acquiring multi-level syntactic translation rules of (Galley et al., 2004) from aligned tree-string pairs, and present two main extensions of their approach: first, instead of merely computing a single derivation that minimally explains a sentence pair, we construct a large number of derivations that include contextually richer rules, and account for multiple interpretations of unaligned words. Second, we propose probability estimates and a training procedure for weighting these rules. We contrast different approaches on real examples, show that our estimates based on multiple derivations favor phrasal re-orderings that are linguistically better motivated, and establish that our larger rules provide a 3.63 BLEU point increase over minimal rules.