YuanheZ/LoRA-One

LoRA-One: One-Step Full Gradient Could Suffice for Fine-Tuning Large Language Models, Provably and Efficiently (ICML2025 Oral)

28
/ 100
Experimental

This project helps machine learning engineers and researchers to fine-tune large language models (LLMs) more efficiently. It takes a pre-trained LLM and a specific dataset, then applies a theoretically-grounded method to quickly adapt the model for new tasks like natural language understanding, mathematical reasoning, or code generation. The output is a fine-tuned LLM that performs better on specialized tasks with significantly less computational effort.

Use this if you are a machine learning practitioner looking for a provably efficient way to adapt large language models to new, specific tasks without extensive full-gradient computations.

Not ideal if you need a solution for models other than large language models or are not familiar with the concepts of fine-tuning and low-rank adaptation.

Large Language Model Fine-tuning Natural Language Processing AI Model Optimization Machine Learning Research Code Generation
No License No Package No Dependents
Maintenance 6 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 7 / 25

How are scores calculated?

Stars

28

Forks

2

Language

Python

License

Category

llm-fine-tuning

Last pushed

Oct 22, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/YuanheZ/LoRA-One"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.