大型语言模型中的拒绝行为远不止一个方向

Computer Science > Computation and Language arXiv:2602.02132 (cs) [Submitted on 2 Feb 2026 (v1), last revised 15 Sep 2026 (this version, v2)] Title:There Is More to Refusal in Large Language Models than a Single Direction Authors:Faaiz Joad, Majd Hawasly, Sabri Boughorbel, Nadir Durrani, Husrev Taha Sencar View a PDF of the paper titled There Is More to Refusal in Large Language Models than a Single Direction, by Faaiz Joad and Majd Hawasly and Sabri Boughorbel and Nadir Durrani and Husrev Taha Sencar View PDF HTML (experimental) Abstract:Prior work argues that refusal in large language models is mediated by a single direction, enabling steering and abliteration. We show that this account is incomplete: across diverse refusal and non-compliance categories, refusal behaviors correspond to geometrically distinct directions in activation space. Yet activation steering along any refusal-related direction produces nearly identical refusal--over-refusal trade-offs, acting as a shared one-dimensional control knob. Thus, different directions primarily affect not whether the model refuses, but how it refuses. Using sparse autoencoders, we uncover a structured internal representation of refusal: a reusable core of shared refusal latents supplemented by style- and domain-specific latents. Linear interventions collapse this structure into uniform behavioral control, flattening mechanistic differences across refusal types. Our results reconcile the apparent simplicity of refusal steering with the diversity of refusal behaviors, and clarify the limits of linear interpretability for aligned model behavior. Comments: Subjects: Computation and Language (cs.CL) Cite as: arXiv:2602.02132 [cs.CL] (or arXiv:2602.02132v2 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2602.02132 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Faaiz Joad [view email] [v1] Mon, 2 Feb 2026 14:15:44 UTC (365 KB) [v2] Tue, 15 Sep 2026 08:32:48 UTC (4,650 KB) Full-text links: Access Paper: View a PDF of the paper titled There Is More to Refusal in Large Language Models than a Single Direction, by Faaiz Joad and Majd Hawasly and Sabri Boughorbel and Nadir Durrani and Husrev Taha Sencar View PDF HTML (experimental) TeX Source view license Current browse context: cs.CL new | recent | 2026-02 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)