RAPS: A Novel Few-Shot Relation Extraction Pipeline with Query-Information Guided Attention and Adaptive Prototype Fusion

TL;DR

This paper introduces RAPS, a new few-shot relation extraction pipeline that uses query-guided attention and adaptive prototype fusion to improve the recognition of unseen relations with limited data.

Contribution

RAPS is the first to integrate query-information guided attention with adaptive prototype fusion for enhanced few-shot relation extraction.

Findings

Significant improvement over state-of-the-art on FewRel 1.0

Effective use of query information for prototype refinement

Adaptive fusion enhances relation representation accuracy

Abstract

Few-shot relation extraction (FSRE) aims at recognizing unseen relations by learning with merely a handful of annotated instances. To generalize to new relations more effectively, this paper proposes a novel pipeline for the FSRE task based on queRy-information guided Attention and adaptive Prototype fuSion, namely RAPS. Specifically, RAPS first derives the relation prototype by the query-information guided attention module, which exploits rich interactive information between the support instances and the query instances, in order to obtain more accurate initial prototype representations. Then RAPS elaborately combines the derived initial prototype with the relation information by the adaptive prototype fusion mechanism to get the integrated prototype for both train and prediction. Experiments on the benchmark dataset FewRel 1.0 show a significant improvement of our method against state-of-the-art methods.