LogicST: A Logical Self-Training Framework for Document-Level Relation Extraction with Incomplete Annotations.

Shengda Fan, Yanting Wang, Shasha Mo, Jianwei Niu

Browse the full EMNLP paper archive.