Demystifying Instruction Mixing for Fine-tuning Large Language Models

Renxi Wang, Haonan Li, Minghao Wu, Yuxia Wang, Xudong Han, Chiyu Zhang, Timothy Baldwin


Abstract
Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore the effects of instruction tuning on different combinations of datasets on LLM performance, and find that certain instruction types are more advantageous for specific applications but can negatively impact other areas. This work provides insights into instruction mixtures, laying the foundations for future research.
Anthology ID:
2024.acl-srw.15
Volume:
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Xiyan Fu, Eve Fleisig
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
68–75
Language:
URL:
https://aclanthology.org/2024.acl-srw.15
DOI:
10.18653/v1/2024.acl-srw.15
Bibkey:
Cite (ACL):
Renxi Wang, Haonan Li, Minghao Wu, Yuxia Wang, Xudong Han, Chiyu Zhang, and Timothy Baldwin. 2024. Demystifying Instruction Mixing for Fine-tuning Large Language Models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop), pages 68–75, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Demystifying Instruction Mixing for Fine-tuning Large Language Models (Wang et al., ACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.acl-srw.15.pdf