Low-Rank Adaptation with Task-Relevant Feature Enhancement for Fine-tuning Language Models

TL;DR

This paper introduces LoRATRF, a method that enhances task-relevant features in low-rank language model fine-tuning, reducing parameters and outperforming state-of-the-art methods across multiple NLP tasks.

Contribution

LoRATRF is a novel approach that incorporates task-aware filtering to improve low-rank adaptation in language models, addressing performance gaps of existing methods.

Findings

Reduces 33.71% parameters compared to SOTA methods.

Achieves better performance on diverse NLP datasets.

Effective across NLU, reasoning, and mathematical tasks.

Abstract

Fine-tuning pre-trained large language models in a parameter-efficient manner is widely studied for its effectiveness and efficiency. LoRA is one of the most widely used methods, which assumes that the optimization process is essentially low dimensional. Although LoRA has demonstrated commendable performance, there remains a significant performance gap between LoRA and full fine-tuning when learning new tasks. In this work, we propose Low-Rank Adaptation with Task-Relevant Feature Enhancement(LoRATRF) for enhancing task-relevant features from the perspective of editing neural network representations. To prioritize task-relevant features, a task-aware filter that selectively extracts valuable knowledge from hidden representations for the target or current task is designed. As the experiments on a vareity of datasets including NLU, commonsense reasoning and mathematical reasoning tasks demonstrates, our method reduces 33.71% parameters and achieves better performance on a variety of datasets in comparison with SOTA low-rank methods.