ReFinED: An Efficient Zero-shot-capable Approach to End-to-End Entity Linking

Tom Ayoola, Shubhi Tyagi, Joseph Fisher, Christos Christodoulopoulos, Andrea Pierleoni
7/8/2022
cs.CL

No official code found. Best implementation: amazon-science/ReFinED (250 stars, Apache-2.0, PyTorch, last commit 2024-12-13).

Abstract

We introduce ReFinED, an efficient end-to-end entity linking model which uses fine-grained entity types and entity descriptions to perform linking. The model performs mention detection, fine-grained entity typing, and entity disambiguation for all mentions within a document in a single forward pass, making it more than 60 times faster than competitive existing approaches. ReFinED also surpasses state-of-the-art performance on standard entity linking datasets by an average of 3.7 F1. The model is capable of generalising to large-scale knowledge bases such as Wikidata (which has 15 times more entities than Wikipedia) and of zero-shot entity linking. The combination of speed, accuracy and scale makes ReFinED an effective and cost-efficient system for extracting entities from web-scale datasets, for which the model has been successfully deployed. Our code and pre-trained models are available at https://github.com/alexa/ReFinED

View on arXivView PDF

Code Implementations(1)

amazon-science/ReFinEDCommunity100%
PyTorchApache-2.0

ReFinED is an efficient and accurate entity linking (EL) system.

25054Shell, PythonMay 3, 20221 years agoApache-2.0
entity-extractionentity-linkingentity-resolutionnlppytorch

Similar Papers

A Read-and-Select Framework for Zero-shot Entity Linking

87% match1 repo92023

Effective Few-Shot Named Entity Linking by Meta-Learning

86% match1 repo42022

Strong Heuristics for Named Entity Linking

86% match1 repo32022

Multilingual End to End Entity Linking

84% match1 repo502023

Improving Few-shot and Zero-shot Entity Linking with Coarse-to-Fine Lexicon-based Retriever

84% match2023

Towards Better Entity Linking with Multi-View Enhanced Distillation

84% match1 repo102023

Cite this paper

@article{ayoola2022refined,
  title  = {ReFinED: An Efficient Zero-shot-capable Approach to End-to-End Entity Linking},
  author = {Tom Ayoola and Shubhi Tyagi and Joseph Fisher and Christos Christodoulopoulos and Andrea Pierleoni},
  year   = {2022},
  eprint = {2207.04108},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2207.04108v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2207.04108.svg)](https://codeofpaper.com/paper/2207.04108)

Discussion