Active Relation Discovery: Towards General and Label-aware OpenRE

Abstract

Open Relation Extraction (OpenRE) aims to discover and label novel relations from open domains. Previous methods mainly suffer from two problems: (1) Insufficient capacity to discriminate between known and novel relations. When extending conventional test settings to a more general setting where test data might also come from seen classes, existing OpenRE approaches have a significant performance decline. (2) Secondary labeling must be performed before practical application. Existing methods cannot label human-readable and meaningful types for novel relations, which is urgently required by the downstream tasks. To address these issues, we propose the Active Relation Discovery (ARD) framework, which utilizes relational outlier detection for discriminating known and novel relations and involves active learning for labeling novel relations. Extensive experimentsThe source code will be available for reproducibility. on three real-world datasets show that ARD significantly outperforms state-of-the-art methods on both conventional and our proposed general OpenRE settings.

Tasks

Reproductions