Activation Steering

Follow this topic Ask about this topic

Activation steering studies how edits to a model’s hidden activations can control its behavior at inference time, enabling changes without retraining the full model. Common approaches add or remove contrastive steering vectors, target sparse autoencoder features, neurons or attention heads, and use conditional or feedback-based controllers to adapt interventions to the input or generation state. Researchers apply these methods to language tasks such as reasoning, refusal, factuality and hallucination mitigation, as well as image and video generation, speech, and embodied agents; evaluations include benchmarks such as GSM8K, TruthfulQA, POPE and LIBERO. A central challenge is that readable or decodable features are not necessarily controllable, and effective edits can vary by context and architecture or cause collateral changes to fluency, utility and safety.

Latest papers

376 papers77 papers in the last four weeks, up 93% on the four weeks before. 0.5% of all new papers.

Mechanistic Personality Analysis of LLMs Steering Personality via Latent Feature Interventions
Jun 27, 2026David Courtis, Ting HuLanguage Model Steering LLM Interpretability
Department of Computing, Queen’s University Kingston, Kingston, ON, Canada.
SPARK: Adaptive Parametric Knowledge Participation for Faithful Retrieval-Augmented Generation
Jun 26, 2026Ruochang Li, Pengcheng Huang, Zhenghao Liu +5Retrieval-Augmented Generation Knowledge Conflicts in Language Models
School of Computer Science and Engineering, Northeastern University, Shenyang, China · Department of Computer Science and Technology, Tsinghua University, Beijing, China
Localizing RL-Induced Tool Use to a Single Crosscoder Feature
Jun 25, 2026Andrii Shportko, Shubham Bhokare, Ahmed Zeyad A Alzahrani +3LLM Interpretability LLM Tool Use
1Northwestern University.
Refusal Lives Downstream of Persona in Chat Models
Jun 24, 2026Viola Zhong, Qirui LiLanguage Model Steering LLM Refusal Behavior
Independent · Department of Mathematics Pohang University of Science and Technology
Detecting and Controlling Sycophancy with Cascading Linear Features
Jun 23, 2026Maty Bohacek, Rishub Jain, Nicholas Dufour +3LLM Sycophancy Language Model Steering
Stanford University · Google DeepMind
Parallel Manifold Steering: Efficient Adaptation of Large Associative Memories via Residual Energy Shaping
Jun 23, 2026Kanishk AwadhiyaMemory-Augmented Neural Networks Transformer Attention
Independent Researcher
Perfect Detection, Failed Control: The Geometry of Knowing vs. Steering in Language Models
Jun 23, 2026Cosimo Galeone, Anna Ettorre, Minsu Park +2Mechanistic Interpretability Causal Interventions in Language Models
Alomana, Grottaglie, Italy
Agent MechSuits: Mechanistic Subspace Safety Steering for Multi-Turn CLI Agents
Jun 21, 2026Weidi Luo, Qiming Zhang, Yihao Quan +5Computer-Use Agents LLM Interpretability
University of Georgia · University of South Florida · Rutgers University +3
ORBIT: Training-Free Multi-Attribute Behavioral Steering via Orthogonal Subspace Rotation
Jun 21, 2026Narges Ghasemi, Amir Ziashahabi, Salman Avestimehr +1Language Model Steering Activation Steering
1Information Sciences Institute, University of Southern California · Department of Electrical and Computer Engineering, University of Southern California
Right Knowledge, Wrong Answer: Test-Time Steering for Temporal Fact Conflicts in Open-Weight Language Models
Jun 18, 2026Elias Hossain, Sourav Saha, Umesh Chandra Biswas +1Knowledge Conflicts in Language Models Temporal Reasoning in Language Models
University of Central Florida · 2Mississippi State University
Translating Inference-Time Control to Radiology Vision-Language Models: Activation Steering for Pneumonia Classification on Chest X-rays
Jun 18, 2026Eduardo Moreno Judice de Mattos Farina, Mateus A. Esmeraldo, Felipe Akio Matsuoka +2Chest X-Ray Classification Radiology VLMs
Universidade Federal de São Paulo (UNIFESP) Hospital Israelita Albert Einstein São Paulo, Brazil · Department of Radiology Stanford University School of Medicine Stanford, CA, USA · Universidade Federal de São Paulo (UNIFESP) DASA São Paulo, Brazil +2
Actionable Activation Directions for Detecting and Mitigating Emergent Misalignment Across Language Model Families
Jun 18, 2026Abdul Rafay SyedLLM Auditing Emergent Misalignment in Language Models
Department of Computer Science Universität des Saarlandes Saarbrücken, Germany
GEMS: Geometric Constraints Enable Multi-Semantic Superposition in LLMs
Jun 18, 2026Yu DengCausal Interventions in Language Models Representation Geometry in Language Models
Closing the Loop: PID Feedback Control for Interpretable Activation Steering in Symbolic Music Generation
Jun 17, 2026Ioannis Prokopiou, Pantelis Vikatos, Maximos Kaliakatsos-Papakostas +2Transformer Interpretability Controllable Music Generation
Athens University of Economics and Business, Athens, Greece · Orfium Research, Athens, Greece · Hellenic Mediterranean University, Chania, Greece +2
Want Better Synthetic Data? Steer It: Activation Steering for Low-Resource Language Generation
Jun 16, 2026Jan Cegin, Daniil Gurgurov, Yusser Al Ghussin +1Language Model Steering Synthetic Data Generation for Language Models
Kempelen Institute of Intelligent Technologies, Bratislava, Slovakia · German Research Institute for Artificial Intelligence (DFKI), Saarbrücken, Germany
SPARK: Security Knowledge Priming and Representation-Guided Knowledge Activation for LLM-based Secure Code Generation
Jun 15, 2026Xiaoyun Xu, Lichao Wu, Jona te Lintelo +2Knowledge Augmentation for Language Models LLM Security
Radboud University · University of Bristol · University of Zagreb
High-Dimensional Random Projection for Activation Steering in Language Models
Jun 13, 2026Minh-Hieu Pham, Bach Do, Laziz Abdullaev +2Language Model Steering Representation Geometry in Language Models
Hanoi University of Science and Technology
Improving Robotic Generalist Policies via Flow Reversal Steering
Jun 11, 2026Andy Tang, William Chen, Andrew Wagenmaker +2RL for Robotics Robot Policy Learning
Stanford University · UC Berkeley
Refusal Beyond a Single Direction: A Preliminary Comparison of Diff-in-Means and INLP
Jun 11, 2026Elisabetta Rocchetti, Alfio FerraraLLM Refusal Behavior Causal Interventions in Language Models
Department of Computer Science, Università degli Studi di Milano, Milan, Italy · Department of Literary Studies, Philology and Linguistics, Università degli Studi di Milano, Milan, Italy
The Signs Were Always There: Training-Free Detection and Control in Raw Transformer Dimensions
Jun 10, 2026Varun Reddy Nalagatla, Madhavi Evana, Yiwen ZhangTransformer Interpretability Representation Engineering
Amazon Web Services
On The Effectiveness-Fluency Trade-Off In LLM Conditioning: A Systematic Study
Jun 10, 2026Iuri Macocco, Pau Rodríguez, Arno Blaas +3LLM Fine-Tuning Activation Steering
Universitat Pompeu Fabra · Apple · Universitat Pompeu Fabra / ICREA
Steering Multirobot Behavior via Closed-Loop Affine Activation Editing
Jun 9, 2026Satyajeet Das, Darren Chiu, Shashank Hegde +1Robot Policy Learning Robot Policy Adaptation
Department of Computer Science University of Southern California Los Angeles, CA, USA
Scenario-based Probing and Steering Cultural Values in Large Language Models--Extended Version
Jun 9, 2026Trung Duc Anh Dang, Tung Kieu, Sarah MasudLanguage Model Steering Cultural Alignment
University of Copenhagen · Aalborg University
Overcoming State Inertia in Full-Duplex Spoken Language Models via Activation Steering
Jun 9, 2026Cheng-Kuang Chang, Kai-Wei Chang, Alexander H. Liu +1Language Model Steering Spoken Dialogue Systems
MIT CSAIL
Recoverable but Not Stationary:Local Linear Structures in Weights and Activations
Jun 9, 2026Irina Piontkovskaia, Sergey NikolenkoLow-Rank Adaptation Activation Steering
St. Petersburg Department of the Steklov Institute of Mathematics, St. Petersburg, Russia · St. Petersburg State University
The Amplifying Mirror: Locating and Steering the Partisan Direction inside a Large Language Model
Jun 7, 2026Wendy K. TamLLM Interpretability Political Bias in Language Models
Professor and Stevenson Chair in the Departments of Political Science, Computer Science, Biomedical Informatics, and the Law School at Vanderbilt University
Activation Steering Induces Emergent Misalignment: Steering Vector Construction and Comprehensive Evaluation
Jun 7, 2026Qi Cao, Jian Lou, Meiting Liu +3Language Model Safety Evaluation LLM Safety
Nanyang Technological University · Sun Yat-sen University · University of Science and Technology of China
Beyond Linear Activation Steering: Invertible Latent Transformations for Controlling LLM Behavior
Jun 7, 2026Tuc Nguyen, Thai LeInvertible Neural Networks Language Model Steering
Indiana University Bloomington, USA
TimpaTeks: Automatic In-place Text Sequence Modification via Diffusion Language Model Steering
Jun 7, 2026Ryandito Diandaru, Ahmed Elshabrawy, Fadli Aulawi Al Ghiffari +2Language Model Steering Diffusion Language Models
MBZUAI
TALAN: Task-Aligned Latent Adaptation Networks for Targeted Post-Training of Large Language Models
Jun 5, 2026Chengkai Zhang, Ziteng Liu, Junpu Wang +4Supervised Fine-Tuning Adapter Tuning
Meta AI

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

Mechanistic Personality Analysis of LLMs Steering Personality via Latent Feature Interventions

David Courtis, Ting Hu

Department of Computing, Queen’s University Kingston, Kingston, ON, Canada.

Language Model Steering LLM Interpretability Personality Modeling in Language Models Causal Interventions in Language Models Activation Steering

Open paper AskPDF arXiv

Large Language Models (LLMs) have demonstrated the ability to simulate human-like OCEAN personality traits in generated text. Previous efforts have focused on prompt engineering or fine-tuning to shape LLM personality. In this work, we propose a mechanistic interpretability approach that directly intervenes on the model's latent features. Our method identifies latent directions in the residual stream corresponding to a target OCEAN trait using sparse autoencoders (SAEs) and contrastive activation analysis. We formalize an additive steering vector in activation space and demonstrate how applying a small additive shift to the hidden states enhances the target trait while preserving overall language modeling performance. To determine the optimal combination of feature shifts, we explore a linear weighting heuristic with grid search optimization that balances personality expression with task performance. Our approach shows promise in controllably steering personality traits at the mechanistic level while maintaining high performance on standard benchmarks.