Publication Type

Conference Proceeding Article

Version

acceptedVersion

Publication Date

8-2024

Abstract

Fine-tuning all parameters of large language models (LLMs) necessitates substantial computational power and extended time. Latest advancements in parameter-efficient fine-tuning (PEFT) techniques, such as Adapter tuning and LoRA, allow for adjustments to only a minor fraction of the parameters of these LLMs. Concurrently, it has been noted that the issue of over-smoothing diminishes the effectiveness of these Transformer-based LLMs, resulting in suboptimal performances in downstream tasks. In this paper, we present SIBO, which is a SImple BOoster to enhance PEFT, by injecting an initial residual. SIBO is straightforward and readily extensible to a range of state-of-the-art PEFT techniques to alleviate over-smoothing and enhance performance. Extensive experiments on 22 benchmark datasets demonstrate that SIBO significantly enhances the performance of various strong baselines, achieving up to 15.7% and 23.5% improvement over existing PEFT methods on the arithmetic and commonsense reasoning tasks, respectively.

Keywords

Large language models, LLMs, Parameter-efficient fine-tuning

Discipline

Artificial Intelligence and Robotics | Computer Sciences

Research Areas

Data Science and Engineering; Intelligent Systems and Optimization

Areas of Excellence

Digital transformation

Publication

62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) : Bangkok, Thailand, August 11-16

First Page

1241

Last Page

1257

Identifier

10.18653/v1/2024.findings-acl.72

Publisher

Association for Computational Linguistics

City or Country

Bangkok, Thailand

Additional URL

https://doi.org/10.18653/v1/2024.findings-acl.72

Download

Share

COinS