Year: 2025
Type: preprint
Abstract: Parameter-Efficient Fine-Tuning (PEFT) methods have become crucial for rapidly adapting large language models (LLMs) to downstream tasks. Prefix-Tuning, an early and effective PEFT technique, demonstrated the ability to achieve performance comparable to full fine-tuning with significantly reduced computational and memory overhead. However, despite its earlier success, its effectiveness in training modern state-of-the-art LLMs has been very limited. In this work, we demonstrate empirically that prefix-tuning underperforms on LLMs because of an inherent tradeoff between the contribution of the input prompt and the parameterized prefix within the attention head. This motivates us to introduce PrefixMemory-Tuning, an architecture that generalizes the principles of prefix-tuning while addressing its shortcomings by shifting the prefix module out of the attention head itself and improving its expressiveness. Our experiments show that, across diverse benchmarks, PrefixMemory-Tuning consistently outperforms existing prefix-tuning methods. Notably, it achieves competitive performance with modern PEFTs on several general benchmarks, highlighting a potential extension of prefix-tuning approaches to become state-of-the-art. Our findings suggest that by overcoming its inherent limitations, prefix-tuning can remain a competitive and relevant research direction in the landscape of parameter-efficient LLM adaptation.
Source: arXiv (Cornell University)
Authors: Wang, Haonan, Chen, Brian, Li, Siquan, Liang, Xinhe, Lee, Hwee Kuan, Kawaguchi, Kenji, Hu, Tianyang
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4415310991
Cites:
Cited by:
Related to: