From Bag of Sentences to Document: Distantly Supervised Relation Extraction via Machine Reading Comprehension
Source code for our paper From Bag of Sentences to Document: Distantly Supervised Relation Extraction via Machine Reading Comprehension
The NYT dataset we used is from THUNLP OpenNRE. We propose it, you can download from here.
Please first uncompress them and put it in the root dir, the path should be like "dataset/NYT/15/xxx"
./train_nyt.sh