arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1702.05053cs.CL
收藏

解决神经AMR解析中的数据稀疏问题

Addressing the Data Sparsity Issue in Neural AMR Parsing

University of Rochester(罗切斯特大学);Brandeis University(布兰迪斯大学)

Xiaochang Peng, Chuan Wang, Daniel Gildea, Nianwen Xue

摘要:

神经注意力模型在不同的NLP任务中取得了巨大的成功。然而,由于数据稀疏性问题,他们并没有履行他们在AMR解析任务上的承诺。在本文中,我们描述了一个序列到序列的AMR解析模型,并提出了不同的方法来解决数据稀疏性问题。我们表明,我们的方法在基线神经注意力模型上取得了显着的改进,我们的结果也与不使用额外语言资源的最先进的系统竞争。

英文摘要:

Neural attention models have achieved great success in different NLP tasks. How- ever, they have not fulfilled their promise on the AMR parsing task due to the data sparsity issue. In this paper, we de- scribe a sequence-to-sequence model for AMR parsing and present different ways to tackle the data sparsity problem. We show that our methods achieve significant improvement over a baseline neural atten- tion model and our results are also compet- itive against state-of-the-art systems that do not use extra linguistic resources.

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。

↑