本文前置知识: BERT: 详见ELMo, GPT, BERT. SpanBERT: Improving Pre-training by Representing and Predicting Spans本文是论文SpanBERT:

本文前置知识: BERT: 详见ELMo, GPT, BERT. StructBERT: Incorporating Language Structures into Pre-training for Deep Language U