QEFT: Quantization for Efficient Fine-Tuning of LLMs
This addresses the problem of efficient fine-tuning for LLM users, offering a practical solution that improves multiple aspects like speed and memory, though it appears incremental as it builds on prior quantization and fine-tuning work.
The paper tackles the challenge of optimizing fine-tuning for large language models (LLMs) by proposing QEFT, a lightweight quantization technique that accelerates both inference and fine-tuning while matching the quality of full-precision methods and using fewer resources.
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.