Abstract
With the rapid growth in the use of fine-tuning for large language models (LLMs), optimizing fine-tuning while keeping inference efficient has become highly important. However, this is a challenging task as it requires improvements in all aspects, including inference speed, fine-tuning speed, memory consumption, and, most importantly, model quality. Previous studies have attempted to achieve this by combining quantization with fine-tuning, but they have failed to enhance all four aspects simultaneously. In this study, we propose a new lightweight technique called Quantization for Efficient Fine-Tuning (QEFT). QEFT accelerates both inference and fine-tuning, is supported by robust theoretical foundations, offers high flexibility, and maintains good hardware compatibility. Our extensive experiments demonstrate that QEFT matches the quality and versatility of full-precision parameter-efficient fine-tuning, while using fewer resources. Our code is available at https://github.com/xvyaward/qeft.- Anthology ID:
- 2024.findings-emnlp.811
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2024
- Month:
- November
- Year:
- 2024
- Address:
- Miami, Florida, USA
- Editors:
- Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 13823–13837
- Language:
- URL:
- https://preview.aclanthology.org/add_missing_videos/2024.findings-emnlp.811/
- DOI:
- 10.18653/v1/2024.findings-emnlp.811
- Cite (ACL):
- Changhun Lee, Jun-gyu Jin, YoungHyun Cho, and Eunhyeok Park. 2024. QEFT: Quantization for Efficient Fine-Tuning of LLMs. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 13823–13837, Miami, Florida, USA. Association for Computational Linguistics.
- Cite (Informal):
- QEFT: Quantization for Efficient Fine-Tuning of LLMs (Lee et al., Findings 2024)
- PDF:
- https://preview.aclanthology.org/add_missing_videos/2024.findings-emnlp.811.pdf