论文题目: Learning Conceptual-Contextual Embeddings for Medical Text
论文摘要:
对于自然语言理解任务来说,外部知识通常是有用的。本文介绍了一个上下文文本表示模型,称为概念上下文(CC)嵌入,它将结构化的知识合并到文本表示中。与实体嵌入方法不同,文中提到的方法将知识图编码到上下文模型中。就像预先训练好的语言模型一样,CC嵌入可以很容易地在广泛的任务中重用。模型利用语义泛化,有效地编码了庞大的UMLS数据库。电子实验健康记录(EHRs)和医疗文本处理基准表明,而使得模型大大提高了监督医疗NLP任务的性能。
A Unitychain is a novel blockchain-like structure that drastically improves transaction scalability and security while maintaining ongoing network performance, even if participating nodes are required to perform a new Distributed Key Generation procedure for security purposes. The Unitychain structure, furthermore, enables greater parallel processing by the assignment of different network node configurations for various database and compute ranges into multiple strands of blockchains that intersect, creating a multi-helix structure, which we call a Unitychain. This thereby enables the network to further bifurcate the roles of nodes into arbitrary yet deterministic network responsibilities in order to maximize the global compute potential.