Language Model Steering

Follow this topic Ask about this topic

Language model steering studies how to make a pretrained model change a particular behavior—such as tone, refusal, truthfulness, reasoning, or language—without necessarily retraining its weights. Common approaches add contrastive activation vectors, target sparse features or specific neurons and attention heads, or adapt interventions to the prompt and the model’s evolving state; other methods guide logits, decoding, or agent-level decisions. Researchers evaluate these methods on tasks ranging from safety and factuality to personalization, multilingual generation, and reasoning, measuring both target behavior and preservation of general capabilities. A central open challenge is reliable, selective control: a behavior may be readable in activations yet resist intervention, while steering can cause collateral quality loss, prompt sensitivity, or safety trade-offs.

Latest papers

283 papers51 papers in the last four weeks, up 65% on the four weeks before. 0.3% of all new papers.

Spokes, not a Ring: Revisiting the Weekday Manifold in Language Models
Oct 27, 2026Diksha Gupta, Rieke Fruengel, Stefan HeimersheimRepresentation Geometry in Language Models LLM Interpretability
Pivotal Research
Orthogonal and Parametrized Multi-Attribute Steering of LLMs
Oct 27, 2026Remy Ogasawara, Cullen Anderson, Narmeen Fatimah Oozeer +4Language Model Steering Activation Steering
Kahlert School of Computing University of Utah Salt Lake City, UT 84112 · Department of Computer Science University of Massachusetts Amherst Amherst, MA 01003 · Martian AI San Francisco, CA 94117, USA +1
CANS: Capability-Preserving Neuron Steering for Large Language Models
Oct 26, 2026Yucheng Wu, Chengyi Xie, Chenpeng Wang +1Language Model Steering Safety-Utility Trade-Off in Language Models
State Key Laboratory of Multimedia Information Processing, Peking University · School of Computer Science, Peking University · YiXin-AILab, YIXIN, Beijing, China
Neyman Orthogonal Activation Steering
Oct 23, 2026David Arbour, Erfan Jahanparast, Serina Chang +1Language Model Steering Activation Steering
Adobe Research · MIT · University of California, Berkeley
DIAL: Steerable Language Models with Distilled Instruction-Adherence Levels
Oct 22, 2026Thanh Q. Tran, John Russell Himawan, Arun Verma +4Language Model Steering Instruction Following
School of Computing, National University of Singapore, Singapore · Singapore-MIT Alliance for Research and Technology Centre, Singapore · CSAIL, Massachusetts Institute of Technology, USA +1
RealControlBench: A Benchmark for Interpretability-Informed Fine-Grained Control of LLMs and Agents
Oct 19, 2026Xinying Cai, Bin Hu, Mingyu Jin +16Language Model Steering Language Model-Based Control
University of California, San Diego · Rutgers University · NEC Laboratories America +8
From Black Box to Controller: Inference-Time Steering of LLM via Feature Amplification
Oct 19, 2026Keqi Chen, Zekai Sun, Yuhua Wen +5Language Model Steering Controllable Text Generation
Beijing University of Posts and Telecommunications, Beijing, China
Compositional Activation Steering Across Languages and Behaviors with Contextual Gates
Oct 17, 2026Daniil Gurgurov, Akansh Maurya, Josef van Genabith +2Language Model Steering Activation Steering
Saarland University · German Research Center for AI (DFKI) · RIKEN +2
What RL Brings to Reasoning: A Few Decisions the Base Model Would Not Make
Oct 17, 2026Siyu Zhang, Yuquan Lu, Yoshua Y. Li +7RL for Language Model Reasoning LLM Reasoning
University of California, San Diego · Meituan, LongCat Team · The Chinese University of Hong Kong, Shenzhen
WAQS: Weight Absorption for Quadratic Probing and Affine Steering
Oct 17, 2026Ding Zhu, Mohammad Mahdi KhaliliLanguage Model Steering Activation Steering
Department of Computer Science and Engineering, the Ohio State University, Columbus, OH 43210, USA
SteerSearch: Interactive Search Reasoning with Mid-Response Steering
Oct 16, 2026Udit Thakur, Baohao Liao, Matteo Nulli +4Language Model Steering Retrieval-Augmented Reasoning
Language Technology Lab, University of Amsterdam · eBay
VISTA: Steering LLMs with Latent User Values Inferred from a Few Survey Answers
Oct 16, 2026Yinuo Xu, Jadie Adams, Brian Hu +3Personalized Language Model Alignment Language Model Steering
University of Michigan · Kitware Inc.
Trajectory-Guided Steering: Mitigating Spurious Feature Bias in Clinical Classification
Oct 15, 2026Fardin Ahsan Sakib, Ziwei Zhu, Ozlem UzunerDebiased ML Language Model Steering
Department of Computer Science, George Mason University · Department of Information Sciences and Technology, George Mason University
Learning to Steer LLM Agents and Multi-Agent Systems
Oct 15, 2026Leyang Shen, Yaxin Gong, Yang Zhang +7LLM Agents Language Model Steering
National University of Singapore · Qwen Business Unit of Alibaba · University of Science and Technology of China
EMO: A Closed-Loop Test-Time Computation Framework via Alternating Context Optimization and Logits Steering
Oct 15, 2026Wei Xia, Yehui Tang, Zhi-Hong DengLanguage Model Steering Large Language Model-Guided Optimization
State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University · Samsung Research, Beijing, China
The Density of Thought: Predicting and Steering Reasoning in Large Language Models
Oct 15, 2026Tanush Kalhan, Anupam Chattopadhyay, Sriram Ranga +1Language Model Steering CoT Reasoning
Plaksha University · Nanyang Technological University · Interview Street Technologies Pvt. Ltd. (HackerRank) +1
Reference-Guided Turn-Taking Style Transfer for Full-Duplex Spoken Language Models
Oct 14, 2026Yuxuan Chen, Yang Xiao, Jack Sun +3Speech Language Models Full-Duplex Spoken Dialogue Systems
The Chinese University of Hong Kong, Shenzhen · The University of Melbourne · Lulula
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations
Oct 14, 2026Enrico Liscio, Anne Arzberger, Aditya Shankar +1LLM Alignment Inference-Time Language Model Alignment
Delft University of Technology, the Netherlands
Multi-System Cognitive Steering for Generative Communication
Oct 13, 2026Eric Lacosse, Pablo Tano, Daniel C. McNameeLanguage Model Steering Language Model-Based Control
Champalimaud Centre for Restorative Neurotechnology, Champalimaud Research, Lisbon, Portugal
Genesis Steering: Better Activation Steering Through Training Data
Oct 13, 2026Tora Bodin, Eric Aubinais, Ilaria Torre +1Language Model Steering Activation Steering
Chalmers University of Technology, Gothenburg, Sweden · Carl von Ossietzky University of Oldenburg, Oldenburg, Germany
Empathy Is Not the Scapegoat for Declines in Factuality and Reliability
Oct 13, 2026Letian Shi, Kanwal Mehreen, Deheng Zhang +1LLM Reliability Language Model Steering
INSAIT, Sofia University “St. Kliment Ohridski” · University of British Columbia
TRACESTEER: Trace Driven Activation Steering for Hallucination Mitigation in Diffusion Language Models
Oct 13, 2026Xin Zhang, Yili Wang, Yue Tan +5Language Model Steering Hallucination in Language Models
Jilin University, China · Griffith University, Australia
Narrative Lookahead in LLMs: Event-Level Decodability, Predictive Advantage, and Steering
Oct 13, 2026Donghoon Jung, Min Kim, Seojin Choi +2Next-Event Prediction Text Generation
School of Digital Humanities and Computational Social Sciences, KAIST, South Korea
The Sharp Directions Are Against You: Curvature Analysis of Activation Steering
Oct 12, 2026Seyedarmin Azizi, Arya Fayyazi, Parsa Razmara +3Language Model Steering Activation Steering
University of Southern California, Los Angeles, USA · Intel AI, USA
FleetPilot: Fine-Grained Steering Empowers Real-World LLM Personalization
Oct 10, 2026Kounianhua Du, Jianxing Liu, Kangning Zhang +7Personalized Language Models Language Model Steering
Shanghai Jiao Tong University · Xiaohongshu Inc., Shanghai, China
Query-Adaptive Neuron Steering for LLMs
Oct 10, 2026Yanxi Pan, Jingwen Wu, Yan Ma +8Language Model Steering
Northeastern University · Washington University in St. Louis · Nanjing University of Aeronautics and Astronautics +4
SteerablePlex: Can We Steer Full-Duplex Models?
Oct 8, 2026Haolong Zheng, Maike Züfle, Dominik Macháček +6Full-Duplex Spoken Dialogue Systems Language Model Steering
University of Illinois Urbana-Champaign · Karlsruhe Institute of Technology · Charles University +3
From Uncertainty to Action: Learning to Steer LLM Agents
Oct 6, 2026Hanwen Li, Jinhao Duan, Guanhua Zhu +4Language Model Steering LLM Agents
New Jersey Institute of Technology · University of North Carolina at Chapel Hill · University of California, Irvine +1
Latent space bias directions in LLMs capture confidence, not fairness
Oct 6, 2026Stephanie Buttigieg, Maeve Madigan, Parameswaran Kamalaruban +1Debiased ML Language Model Steering
Institute of Astronomy & Kavli Institute for Cosmology, University of Cambridge, UK · Risk and Security AI Lab, Visa Inc., Cambridge, UK
Does Steering Break Your Model? A Multi-Dimensional Evaluation Suite for LLM Steering Methods
Oct 6, 2026Haotian Yang, Huikang Jiang, Yucheng Wu +4LLM Evaluation Language Model Steering
State Key Laboratory of Multimedia Information Processing, Peking University · School of Computer Science, Peking University · Columbia University +2

Oct 27, 2026J/K move · Enter open · S save

Spokes, not a Ring: Revisiting the Weekday Manifold in Language Models

Diksha Gupta, Rieke Fruengel, Stefan Heimersheim

Pivotal Research

Representation Geometry in Language Models LLM Interpretability Language Model Steering

Open paper AskPDF arXiv

Recent work has hypothesised that large language models may represent some concepts as multi-dimensional manifolds instead of single directions. Using these manifolds to interpret or steer representations often assumes that moving along them corresponds to a continuous change in the represented concept. In this work, we test this continuous manifold assumption on the prototypical weekday manifold. Despite its striking circular geometry, we find that the weekday representations behave more like seven discrete states arranged in a circle than a continuous ring. First, we show that local sensitivity has seven minima aligned with the weekdays, producing a step-like mapping between representation and behaviour under on-manifold steering. Second, we test whether any concepts are represented along the manifold outside the seven discrete states, and find neither time of day, uncertainty, nor natural text populate the intermediate space. Together, these results challenge the continuous interpretation of the weekday manifold and motivate testing whether movement along other representation manifolds correspond to meaningful changes in model representations and behaviour.