arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1604.03035cs.CL
收藏

基于全局特征的共指消解

Learning Global Features for Coreference Resolution

School of Engineering and Applied Sciences(工程与应用科学学院);Harvard University(哈佛大学)

Sam Wiseman, Alexander M. Rush, Stuart M. Shieber

摘要:

有令人信服的证据表明,共指预测将受益于实体集群的全球信息建模。然而,国家的最先进的性能可以实现与系统处理每个提到预测独立,我们归因于固有的困难,制作信息集群级功能。相反,我们建议使用递归神经网络(RNN)来直接从实体集群的提及中学习它们的潜在全局表示。我们表明,这种表示是特别有用的代词提及的预测,并可以纳入一个端到端的共指系统,优于最先进的,而不需要任何额外的搜索。

英文摘要:

There is compelling evidence that coreference prediction would benefit from modeling global information about entity-clusters. Yet, state-of-the-art performance can be achieved with systems treating each mention prediction independently, which we attribute to the inherent difficulty of crafting informative cluster-level features. We instead propose to use recurrent neural networks (RNNs) to learn latent, global representations of entity clusters directly from their mentions. We show that such representations are especially useful for the prediction of pronominal mentions, and can be incorporated into an end-to-end coreference system that outperforms the state of the art without requiring any additional search.

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。

↑