Skip to content

This is the repository for ERAlign, the proposed framework of "Unsupervised Robust Cross-Lingual Entity Alignment via Joint Modeling of Entity and Relation Texts" (WSDM 2025).

Notifications You must be signed in to change notification settings

SoojinY/ERAlign

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 

Repository files navigation

ERAlign

This is the repository for ERAlign, the proposed framework of "Unsupervised Robust Cross-Lingual Entity Alignment via Joint Modeling of Entity and Relation Texts" (WSDM 2025).

Our paper is available at arXiv: https://arxiv.org/abs/2407.15588

Requirement

  • python == 3.9
  • numpy == 1.24.3
  • tensorflow == 2.13.0
  • torch == 2.0.1
  • tqdm
  • sentence-transformers == 2.2.2

Code

To run with default setting, please use : python main.py

Datasets

You can download the dataset here.

These datasets are provided by:

Zequn Sun, Wei Hu, and Chengkai Li. 2017. Cross-Lingual Entity Alignment via Joint Attribute-Preserving Embedding. In The Semantic Web – ISWC 2017

Citation

If our work was helpful for your project, cite our work :)

@inproceedings{10.1145/3701551.3703500,
author = {Yoon, Soojin and Ko, Sungho and Kim, Tongyoung and Kang, SeongKu and Yeo, Jinyoung and Lee, Dongha},
title = {Unsupervised Robust Cross-Lingual Entity Alignment via Neighbor Triple Matching with Entity and Relation Texts},
year = {2025},
isbn = {9798400713293},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3701551.3703500},
doi = {10.1145/3701551.3703500},
booktitle = {Proceedings of the Eighteenth ACM International Conference on Web Search and Data Mining},
pages = {184–193},
numpages = {10},
keywords = {cross-lingual entity alignment, knowledge graph, neighbor triple matching, optimal transport, pretrained language models},
location = {Hannover, Germany},
series = {WSDM '25}
}

About

This is the repository for ERAlign, the proposed framework of "Unsupervised Robust Cross-Lingual Entity Alignment via Joint Modeling of Entity and Relation Texts" (WSDM 2025).

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages