Language Embeddings for Typology and Cross-lingual Transfer Learning

Trained language embeddings and scripts for the paper:

Language Embeddings for Typology and Cross-lingual Transfer Learning by Dian Yu, Taiqi He and Kenji Sagae. ACL 2021

Abstract

Cross-lingual language tasks typically requirea substantial amount of annotated data or par-allel translation data.We explore whetherlanguage representations that capture relation-ships among languages can be learned andsubsequently leveraged in cross-lingual taskswithout the use of parallel data. We gener-ate dense embeddings for 29 languages usinga denoising autoencoder, and evaluate the em-beddings using the World Atlas of LanguageStructures (WALS) and two extrinsic tasks ina zero-shot setting: cross-lingual dependencyparsing and cross-lingual natural language in-ference.

Bibtex:

@inproceedings{yu-etal-2021-language,
    title = "Language Embeddings for Typology and Cross-lingual Transfer Learning",
    author = "Yu, Dian  and
      He, Taiqi  and
      Sagae, Kenji",
    booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.acl-long.560",
    doi = "10.18653/v1/2021.acl-long.560",
    pages = "7210--7225",
}