AI Hub
All papers
Training

LoRA: Low-Rank Adaptation of Large Language Models

Microsoft·June 17, 2021

Edward J. Hu, Yelong Shen, Phillip Wallis

View on arXiv

TL;DR

Introduces LoRA: freeze the base model and train small low-rank adapter matrices, cutting fine-tuning cost by orders of magnitude with no quality loss.

Why it matters

Made fine-tuning large models affordable and democratized customization. LoRA and its descendants underpin most of the open-source fine-tuning ecosystem.

Related terms