Collaborative and Efficient Fine-tuning: Leveraging Task Similarity

TL;DR

This paper introduces CoLoRA, a collaborative fine-tuning method that leverages task similarity to improve the efficiency and effectiveness of adapting foundation models to multiple related downstream tasks.

Contribution

The paper proposes CoLoRA, a novel approach that uses shared and personalized adapters to exploit task similarity, with theoretical guarantees and empirical validation.

Findings

CoLoRA significantly improves performance when fine-tuning with similar tasks.

Theoretical guarantees for ground truth recovery in linear regression.

Empirical results show boosted individual performance with task collaboration.

Abstract

Adaptability has been regarded as a central feature in the foundation models, enabling them to effectively acclimate to unseen downstream tasks. Parameter-efficient fine-tuning methods such as celebrated LoRA facilitate efficient adaptation of large foundation models using labeled, high-quality and generally scarce task data. To mitigate data scarcity in fine-tuning of foundation models, we propose to leverage task similarity across multiple downstream users. Intuitively, users with similar tasks must be able to assist each other in boosting the effective fine-tuning data size. We propose Collaborative Low-Rank Adaptation, or CoLoRA, which exploits task similarity to collaboratively and efficiently fine-tune personalized foundation models. The main idea in CoLoRA is to train one shared adapter capturing underlying task similarities across all tasks, and personalized adapters tailored to user-specific tasks. We theoretically study CoLoRA on heterogeneous linear regression and provide provable guarantees for ground truth recovery. We also conduct several natural language experiments with varying task similarity, which further demonstrate that when trained together with similar tasks, individual performances are significantly boosted.