MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learning

TL;DR

MTLoRA introduces a low-rank adaptation framework for multi-task learning that significantly reduces trainable parameters while improving accuracy, enabling efficient and effective multi-task model fine-tuning.

Contribution

The paper presents MTLoRA, a novel parameter-efficient multi-task learning framework using low-rank modules, addressing the gap in multi-task adaptation methods.

Findings

Achieves higher accuracy than full fine-tuning on PASCAL tasks.

Reduces trainable parameters by 3.6 times compared to full fine-tuning.

Outperforms existing parameter-efficient methods in accuracy and efficiency.

Abstract

Adapting models pre-trained on large-scale datasets to a variety of downstream tasks is a common strategy in deep learning. Consequently, parameter-efficient fine-tuning methods have emerged as a promising way to adapt pre-trained models to different tasks while training only a minimal number of parameters. While most of these methods are designed for single-task adaptation, parameter-efficient training in Multi-Task Learning (MTL) architectures is still unexplored. In this paper, we introduce MTLoRA, a novel framework for parameter-efficient training of MTL models. MTLoRA employs Task-Agnostic and Task-Specific Low-Rank Adaptation modules, which effectively disentangle the parameter space in MTL fine-tuning, thereby enabling the model to adeptly handle both task specialization and interaction within MTL contexts. We applied MTLoRA to hierarchical-transformer-based MTL architectures, adapting them to multiple downstream dense prediction tasks. Our extensive experiments on the PASCAL dataset show that MTLoRA achieves higher accuracy on downstream tasks compared to fully fine-tuning the MTL model while reducing the number of trainable parameters by 3.6x. Furthermore, MTLoRA establishes a Pareto-optimal trade-off between the number of trainable parameters and the accuracy of the downstream tasks, outperforming current state-of-the-art parameter-efficient training methods in both accuracy and efficiency. Our code is publicly available.