SpanBERT: Improving Pre-training by Representing and Predicting Spans.

Joshi Mandar,Chen Danqi,Liu Yinhan,Weld Daniel S.,Zettlemoyer Luke,Levy Omer

Trans. Assoc. Comput. Linguistics（2020）

引用 1975|浏览821

暂无评分

摘要

We present SpanBERT, a pre-training method that is designed to better represent and predict spans of text. Our approach extends BERT by (1) masking contiguous random spans, rather than random token...

查看译文

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要