Few-shot Open Relation Extraction with Gaussian Prototype and Adaptive Margin

TL;DR

This paper introduces GPAM, a novel framework for few-shot open relation extraction that uses Gaussian prototypes and adaptive margins to improve classification boundaries and handle unknown classes more effectively.

Contribution

The paper proposes a new Gaussian prototype and adaptive margin framework (GPAM) that enhances few-shot relation extraction with better representations and boundary learning.

Findings

GPAM outperforms previous methods on FewRel dataset.

Achieves state-of-the-art accuracy in few-shot open relation extraction.

Demonstrates robustness and stability through contrastive learning.

Abstract

Few-shot relation extraction with none-of-the-above (FsRE with NOTA) aims at predicting labels in few-shot scenarios with unknown classes. FsRE with NOTA is more challenging than the conventional few-shot relation extraction task, since the boundaries of unknown classes are complex and difficult to learn. Meta-learning based methods, especially prototype-based methods, are the mainstream solutions to this task. They obtain the classification boundary by learning the sample distribution of each class. However, their performance is limited because few-shot overfitting and NOTA boundary confusion lead to misclassification between known and unknown classes. To this end, we propose a novel framework based on Gaussian prototype and adaptive margin named GPAM for FsRE with NOTA, which includes three modules, semi-factual representation, GMM-prototype metric learning and decision boundary learning. The first two modules obtain better representations to solve the few-shot problem through debiased information enhancement and Gaussian space distance measurement. The third module learns more accurate classification boundaries and prototypes through adaptive margin and negative sampling. In the training procedure of GPAM, we use contrastive learning loss to comprehensively consider the effects of range and margin on the classification of known and unknown classes to ensure the model's stability and robustness. Sufficient experiments and ablations on the FewRel dataset show that GPAM surpasses previous prototype methods and achieves state-of-the-art performance.