Open Yuanlim0919 opened 1 year ago
您好,在阅读paper的时候paper有提到说这个模型error的主要来源是document retrieval和sentence selection这两个步骤。想请问如果能够将大量的Wikipedia document连接成一个knowledge graph 是否有助于提升模型的accuracy呢?(目前手上有约100万笔的维基条目资料)谢谢!
您好,在阅读paper的时候paper有提到说这个模型error的主要来源是document retrieval和sentence selection这两个步骤。想请问如果能够将大量的Wikipedia document连接成一个knowledge graph 是否有助于提升模型的accuracy呢?(目前手上有约100万笔的维基条目资料)谢谢!