arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.04625cs.CLcs.AIcs.LG
收藏

神经语言模型中的成分表征--以并列短语为例

Representation of Constituents in Neural Language Models: Coordination Phrase as a Case Study

Université de Paris(巴黎大学);CNRS(法国国家科学研究中心);MIT(麻省理工学院);Harvard University(哈佛大学)

Aixiu An, Peng Qian, Ethan Wilcox, Roger Levy

摘要:

神经语言模型已经在许多NLP任务上取得了最先进的性能,最近已经被证明可以学习单个单词之间的许多层次敏感的句法依赖关系。然而,对于语言处理来说,同样重要的是将单词组合成短语成分的能力,以及使用成分级特征来驱动下游期望的能力。在这里,我们研究神经模型的能力,代表成分水平的功能,使用协调名词短语作为案例研究。我们评估在英语和法语上训练的不同神经语言模型是否代表短语级别的数字和性别特征,并使用这些特征来驱动下游期望。我们的研究结果表明,模型使用NP成分数的线性组合来驱动CoordNP/动词数协议。这种行为是高度规则的,甚至对局部句法上下文敏感,但它与观察到的人类行为有很大的不同。模型在性别协议方面的成功率较低。在大型语料库上训练的模型表现最好,使用显式语法监督训练的模型没有明显的优势。

英文摘要:

Neural language models have achieved state-of-the-art performances on many NLP tasks, and recently have been shown to learn a number of hierarchically-sensitive syntactic dependencies between individual words. However, equally important for language processing is the ability to combine words into phrasal constituents, and use constituent-level features to drive downstream expectations. Here we investigate neural models' ability to represent constituent-level features, using coordinated noun phrases as a case study. We assess whether different neural language models trained on English and French represent phrase-level number and gender features, and use those features to drive downstream expectations. Our results suggest that models use a linear combination of NP constituent number to drive CoordNP/verb number agreement. This behavior is highly regular and even sensitive to local syntactic context, however it differs crucially from observed human behavior. Models have less success with gender agreement. Models trained on large corpora perform best, and there is no obvious advantage for models trained using explicit syntactic supervision.

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。

↑