Controllable Text Simplification with Lexical Constraint Loss
Controllable Text Simplification with Lexical Constraint Loss
复制标题
DOI：
10.18653/v1/p19-2036
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Daiki Nishihara;Tomoyuki Kajiwara;Yuki Arase
中科院分区：
文献类型：
--
作者：
Daiki Nishihara;Tomoyuki Kajiwara;Yuki Arase
本文提出了一种在文本简化任务中控制句子层次的方法。文本简化是一种单语翻译任务，将复杂的句子翻译成更简单、更容易理解的替代方案。在这项研究中，我们使用美国教育系统的年级水平作为句子的水平。我们的文本简化方法通过考虑句子和单词的水平，成功地将输入翻译成特定的年级水平。句子级别通过添加目标年级作为输入来考虑。相比之下，通过基于频繁出现在期望等级水平的句子中的单词向训练损失添加权重来考虑单词水平。虽然现有的模型只考虑句子水平可以控制句法复杂度，他们往往会产生超出目标水平的单词。我们的方法可以控制词汇和句法的复杂性，并实现积极的重写。实验结果表明，该方法提高了BLEU和SARI的度量。
We propose a method to control the level of a sentence in a text simplification task. Text simplification is a monolingual translation task translating a complex sentence into a simpler and easier to understand the alternative. In this study, we use the grade level of the US education system as the level of the sentence. Our text simplification method succeeds in translating an input into a specific grade level by considering levels of both sentences and words. Sentence level is considered by adding the target grade level as input. By contrast, the word level is considered by adding weights to the training loss based on words that frequently appear in sentences of the desired grade level. Although existing models that consider only the sentence level may control the syntactic complexity, they tend to generate words beyond the target level. Our approach can control both the lexical and syntactic complexity and achieve an aggressive rewriting. Experiment results indicate that the proposed method improves the metrics of both BLEU and SARI.