The Fine-Tuning Paradox: Boosting Translation Quality Without Sacrificing LLM Abilities

David Stap, Eva Hasler, Bill Byrne, Christof Monz, Ke Tran


Abstract
Fine-tuning large language models (LLMs) for machine translation has shown improvements in overall translation quality. However, it is unclear what is the impact of fine-tuning on desirable LLM behaviors that are not present in neural machine translation models, such as steerability, inherent document-level translation abilities, and the ability to produce less literal translations. We perform an extensive translation evaluation on the LLaMA and Falcon family of models with model size ranging from 7 billion up to 65 billion parameters.Our results show that while fine-tuning improves the general translation quality of LLMs, several abilities degrade. In particular, we observe a decline in the ability to perform formality steering, to produce technical translations through few-shot examples, and to perform document-level translation. On the other hand, we observe that the model produces less literal translations after fine-tuning on parallel data. We show that by including monolingual data as part of the fine-tuning data we can maintain the abilities while simultaneously enhancing overall translation quality. Our findings emphasize the need for fine-tuning strategies that preserve the benefits of LLMs for machine translation.
Anthology ID:
2024.acl-long.336
Volume:
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6189–6206
Language:
URL:
https://aclanthology.org/2024.acl-long.336
DOI:
10.18653/v1/2024.acl-long.336
Bibkey:
Cite (ACL):
David Stap, Eva Hasler, Bill Byrne, Christof Monz, and Ke Tran. 2024. The Fine-Tuning Paradox: Boosting Translation Quality Without Sacrificing LLM Abilities. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6189–6206, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
The Fine-Tuning Paradox: Boosting Translation Quality Without Sacrificing LLM Abilities (Stap et al., ACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.acl-long.336.pdf