Enhancing Data Quality in Federated Fine-Tuning of Foundation Models

TL;DR

This paper introduces a data quality control pipeline for federated fine-tuning of foundation models, enabling collaboration across private data sources while maintaining data privacy and improving overall model performance.

Contribution

It proposes a novel data quality scoring and thresholding method tailored for federated learning, enhancing data selection and model training effectiveness.

Findings

Improved model performance with quality-controlled data

Enhanced reliability of federated fine-tuning process

Effective data scoring and thresholding mechanism

Abstract

In the current landscape of foundation model training, there is a significant reliance on public domain data, which is nearing exhaustion according to recent research. To further scale up, it is crucial to incorporate collaboration among multiple specialized and high-quality private domain data sources. However, the challenge of training models locally without sharing private data presents numerous obstacles in data quality control. To tackle this issue, we propose a data quality control pipeline for federated fine-tuning of foundation models. This pipeline computes scores reflecting the quality of training data and determines a global threshold for a unified standard, aiming for improved global performance. Our experiments show that the proposed quality control pipeline facilitates the effectiveness and reliability of the model training, leading to better performance.