Open Set Relation Extraction via Unknown-Aware Training

by   Jun Zhao, et al.
FUDAN University

The existing supervised relation extraction methods have achieved impressive performance in a closed-set setting, where the relations during both training and testing remain the same. In a more realistic open-set setting, unknown relations may appear in the test set. Due to the lack of supervision signals from unknown relations, a well-performing closed-set relation extractor can still confidently misclassify them into known relations. In this paper, we propose an unknown-aware training method, regularizing the model by dynamically synthesizing negative instances. To facilitate a compact decision boundary, “difficult” negative instances are necessary. Inspired by text adversarial attacks, we adaptively apply small but critical perturbations to original training instances and thus synthesizing negative instances that are more likely to be mistaken by the model as known relations. Experimental results show that this method achieves SOTA unknown relation detection without compromising the classification of known relations.


page 1

page 2

page 3

page 4


H-FND: Hierarchical False-Negative Denoising for Distant Supervision Relation Extraction

Although distant supervision automatically generates training data for r...

RESIDE: Improving Distantly-Supervised Neural Relation Extraction using Side Information

Distantly-supervised Relation Extraction (RE) methods train an extractor...

Actively Supervised Clustering for Open Relation Extraction

Current clustering-based Open Relation Extraction (OpenRE) methods usual...

STAD: Self-Training with Ambiguous Data for Low-Resource Relation Extraction

We present a simple yet effective self-training approach, named as STAD,...

False Positive and Cross-relation Signals in Distant Supervision Data

Distant supervision (DS) is a well-established method for relation extra...

Active Relation Discovery: Towards General and Label-aware Open Relation Extraction

Open Relation Extraction (OpenRE) aims to discover novel relations from ...

Please sign up or login with your details

Forgot password? Click here to reset