Improving Long Tailed Document-Level Relation Extraction via Easy Relation Augmentation and Contrastive Learning

21 May 2022  ·  Yangkai Du, Tengfei Ma, Lingfei Wu, Yiming Wu, Xuhong Zhang, Bo Long, Shouling Ji ·

Towards real-world information extraction scenario, research of relation extraction is advancing to document-level relation extraction(DocRE). Existing approaches for DocRE aim to extract relation by encoding various information sources in the long context by novel model architectures. However, the inherent long-tailed distribution problem of DocRE is overlooked by prior work. We argue that mitigating the long-tailed distribution problem is crucial for DocRE in the real-world scenario. Motivated by the long-tailed distribution problem, we propose an Easy Relation Augmentation(ERA) method for improving DocRE by enhancing the performance of tailed relations. In addition, we further propose a novel contrastive learning framework based on our ERA, i.e., ERACL, which can further improve the model performance on tailed relations and achieve competitive overall DocRE performance compared to the state-of-arts.

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Relation Extraction DocRED ERACL F1 61.36 # 25
Ign F1 59.08 # 27