Abstract
Adapting English-based large language models (LLMs) to other languages has become increasingly popular due to the efficiency and potential of cross-lingual transfer. However, existing language adaptation methods often overlook the benefits of cross-lingual supervision. In this study, we introduce LEIA, a language adaptation tuning method that utilizes Wikipedia entity names aligned across languages. This method involves augmenting the target language corpus with English entity names and training the model using left-to-right language modeling. We assess LEIA on diverse question answering datasets using 7B-parameter LLMs, demonstrating significant performance gains across various non-English languages.- Anthology ID:
- 2024.findings-acl.419
- Volume:
- Findings of the Association for Computational Linguistics: ACL 2024
- Month:
- August
- Year:
- 2024
- Address:
- Bangkok, Thailand
- Editors:
- Lun-Wei Ku, Andre Martins, Vivek Srikumar
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 7029–7039
- Language:
- URL:
- https://preview.aclanthology.org/build-pipeline-with-new-library/2024.findings-acl.419/
- DOI:
- 10.18653/v1/2024.findings-acl.419
- Cite (ACL):
- Ikuya Yamada and Ryokan Ri. 2024. LEIA: Facilitating Cross-lingual Knowledge Transfer in Language Models with Entity-based Data Augmentation. In Findings of the Association for Computational Linguistics: ACL 2024, pages 7029–7039, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- LEIA: Facilitating Cross-lingual Knowledge Transfer in Language Models with Entity-based Data Augmentation (Yamada & Ri, Findings 2024)
- PDF:
- https://preview.aclanthology.org/build-pipeline-with-new-library/2024.findings-acl.419.pdf