[2603.03324] Controlling Chat Style in Language Models via Single-Direction Editing

Computer Science > Computation and Language arXiv:2603.03324 (cs) [Submitted on 10 Feb 2026] Title:Controlling Chat Style in Language Models via Single-Direction Editing Authors:Zhenyu Xu, Victor S. Sheng View a PDF of the paper titled Controlling Chat Style in Language Models via Single-Direction Editing, by Zhenyu Xu and Victor S. Sheng View PDF HTML (experimental) Abstract: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. Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2603.03324 [cs.CL]   (or arXiv:2603.03324v1 [cs.CL] for this version)   https://doi.org/10.48550/arXiv.2603.03324 Focus to learn more arXiv-issued DOI via...