Customizing Large Language Model Generation Style using Parameter-Efficient Finetuning

Xinyue Liu, Harshita Diddee, Daphne Ippolito
9/6/2024
cs.CL

No official code found. Best implementation: cauchy221/StyleTunedLM (4 stars, last commit 2025-01-15).

Abstract

One-size-fits-all large language models (LLMs) are increasingly being used to help people with their writing. However, the style these models are trained to write in may not suit all users or use cases. LLMs would be more useful as writing assistants if their idiolect could be customized to match each user. In this paper, we explore whether parameter-efficient finetuning (PEFT) with Low-Rank Adaptation can effectively guide the style of LLM generations. We use this method to customize LLaMA-2 to ten different authors and show that the generated text has lexical, syntactic, and surface alignment with the target author but struggles with content memorization. Our findings highlight the potential of PEFT to support efficient, user-level customization of LLMs.

View on arXivView PDF

Code Implementations(1)

cauchy221/StyleTunedLMCommunity64%

The official repository of the paper: Customizing Large Language Model Generation Style using Parameter-Efficient Finetuning

42PythonJan 15, 20251 years ago

Similar Papers

An Empirical Study on Parameter-Efficient Fine-Tuning for MultiModal Large Language Models

86% match1 repo252024

Can Smaller LLMs do better? Unlocking Cross-Domain Potential through Parameter-Efficient Fine-Tuning for Text Summarization

86% match2025

Democratizing Large Language Models via Personalized Parameter-Efficient Fine-tuning

86% match1 repo512024

Simplify-This: A Comparative Analysis of Prompt-Based and Fine-Tuned LLMs

85% match1 repo32026

Small Language Models Improve Giants by Rewriting Their Outputs

85% match1 repo122023

Are LLMs Effective Backbones for Fine-tuning? An Experimental Investigation of Supervised LLMs on Chinese Short Text Matching

85% match2024

Cite this paper

@article{liu2024customizing,
  title  = {Customizing Large Language Model Generation Style using Parameter-Efficient Finetuning},
  author = {Xinyue Liu and Harshita Diddee and Daphne Ippolito},
  year   = {2024},
  eprint = {2409.04574},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.04574v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2409.04574.svg)](https://codeofpaper.com/paper/2409.04574)

Discussion