Relation Classification as Two-way Span-Prediction

9 Oct 2020  ·  Amir DN Cohen, Shachar Rosenman, Yoav Goldberg ·

The current supervised relation classification (RC) task uses a single embedding to represent the relation between a pair of entities. We argue that a better approach is to treat the RC task as span-prediction (SP) problem, similar to Question answering (QA). We present a span-prediction based system for RC and evaluate its performance compared to the embedding based system. We demonstrate that the supervised SP objective works significantly better then the standard classification based objective. We achieve state-of-the-art results on the TACRED and SemEval task 8 datasets.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Relation Extraction SemEval-2010 Task-8 SP F1 91.9 # 1
Relation Extraction TACRED Relation Reduction F1 74.8 # 8

Methods


No methods listed for this paper. Add relevant methods here