Controlling stylistic attributes in large language models (LLMs) remains challenging, with existing approaches relying on either prompt engineering or post-training alignment. This paper investigates this challenge through the lens of representation engineering, testing the hypothesis that distinct stylistic attributes - from emotional tone to linguistic structure - are encoded as linear directions in the model's activation space. We provide strong empirical evidence for this hypothesis across a wide range of styles and, based on this finding, present a lightweight, training-free method for precise style control. Our approach supports linear style composition, enhances safety by ablating undesirable behaviors, and, as confirmed by experiments on over a dozen models, achieves high style adherence while preserving core capabilities at minimal computational cost.
@article{arxiv.2603.03324,
title = {Controlling Chat Style in Language Models via Single-Direction Editing},
author = {Zhenyu Xu and Victor S. Sheng},
journal= {arXiv preprint arXiv:2603.03324},
year = {2026}
}