DARE: Data Augmented Relation Extraction with GPT-2

Yannis Papanikolaou, Andrea Pierleoni

Keywords: GPT-2, Relation Extraction, NLP, BERT

TL;DR: Data Augmented Relation Extraction with GPT-2

Abstract: Real-world Relation Extraction (RE) tasks are challenging to deal with, either due to limited training data or class imbalance issues. In this work, we present Data Augmented Relation Extraction (DARE), a simple method to augment training data by properly finetuning GPT2 to generate examples for specific relation types. The generated training data is then used in combination with the gold dataset to train a BERT-based RE classifier. In a series of experiments we show the advantages of our method, which leads in improvements of up to 11 F1 score points compared to a strong baseline. Also, DARE achieves new state-of-the-art in three widely used biomedical RE datasets surpassing the previous best results by 4.7 F1 points on average.