GLiREL -- Generalist Model for Zero-Shot Relation Extraction

TL;DR

GLiREL is a new lightweight model that efficiently performs zero-shot relation extraction, achieving state-of-the-art results and introducing a synthetic dataset generation protocol.

Contribution

It presents a novel architecture and training paradigm for zero-shot relation classification, inspired by zero-shot NER advancements.

Findings

Achieves state-of-the-art results on FewRel and WikiZSL benchmarks.

Introduces a protocol for synthetic dataset generation with diverse relation labels.

Demonstrates efficient and accurate zero-shot relation prediction in a single forward pass.

Abstract

We introduce GLiREL (Generalist Lightweight model for zero-shot Relation Extraction), an efficient architecture and training paradigm for zero-shot relation classification. Inspired by recent advancements in zero-shot named entity recognition, this work presents an approach to efficiently and accurately predict zero-shot relationship labels between multiple entities in a single forward pass. Experiments using the FewRel and WikiZSL benchmarks demonstrate that our approach achieves state-of-the-art results on the zero-shot relation classification task. In addition, we contribute a protocol for synthetically-generating datasets with diverse relation labels.