Breaking the Curse of Multilinguality with Cross-lingual Expert Language Models
Terra Blevins, Tomasz Limisiewicz, Suchin Gururangan, Margaret Li, Hila Gonen, Noah A. Smith, Luke Zettlemoyer
Abstract
Despite their popularity in non-English NLP, multilingual language models often underperform monolingual ones due to inter-language competition for model parameters. We propose Cross-lingual Expert Language Models (X-ELM), which mitigate this competition by independently training language models on subsets of the multilingual corpus. This process specializes X-ELMs to different languages while remaining effective as a multilingual ensemble. Our experiments show that when given the same compute budget, X-ELM outperforms jointly trained multilingual models across all 16 considered languages and that these gains transfer to downstream tasks. X-ELM provides additional benefits over performance improvements: new experts can be iteratively added, adapting X-ELM to new languages without catastrophic forgetting. Furthermore, training is asynchronous, reducing the hardware requirements for multilingual training and democratizing multilingual modeling.
BibTeX
@inproceedings{blevins-etal-2024-breaking,
title = "Breaking the Curse of Multilinguality with Cross-lingual Expert Language Models",
author = "Blevins, Terra and
Limisiewicz, Tomasz and
Gururangan, Suchin and
Li, Margaret and
Gonen, Hila and
Smith, Noah A. and
Zettlemoyer, Luke",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.604/",
doi = "10.18653/v1/2024.emnlp-main.604",
pages = "10822--10837"
}