Yuan Yao, Deming Ye, Peng Li +7
The paper introduces DocRED, a large human‑annotated dataset for extracting relations between entities across multiple sentences in a document, and provides both supervised and dis…