Cover Image

Extracting Unlearned Information from LLMs with Activation Steering

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Seyitoğlu, Atakan, Kuvshinov, Aleksei, Schwinn, Leo, Günnemann, Stephan
Format:	Preprint
Published:	2024
Subjects:	Computation and Language Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Intriguing Properties of Input-dependent Randomized Smoothing
by: Súkeník, Peter, et al.
Published: (2021)

A Probabilistic Perspective on Unlearning and Alignment for Large Language Models
by: Scholten, Yan, et al.
Published: (2024)

Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs
by: Scholten, Yan, et al.
Published: (2025)

Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting
by: Kollovieh, Marcel, et al.
Published: (2024)

HyperSteer: Activation Steering at Scale with Hypernetworks
by: Sun, Jiuding, et al.
Published: (2025)

Steer Like the LLM: Activation Steering that Mimics Prompting
by: Heyman, Geert, et al.
Published: (2026)

Assessing Robustness via Score-Based Adversarial Image Generation
by: Kollovieh, Marcel, et al.
Published: (2023)

A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness
by: Schwinn, Leo, et al.
Published: (2026)

Programming Refusal with Conditional Activation Steering
by: Lee, Bruce W., et al.
Published: (2024)

SAKE: Steering Activations for Knowledge Editing
by: Scialanga, Marco, et al.
Published: (2025)

Closing the Distribution Gap in Adversarial Training for LLMs
by: Hu, Chengzhi, et al.
Published: (2026)

Endogenous Resistance to Activation Steering in Language Models
by: McKenzie, Alex, et al.
Published: (2026)

Tool Unlearning for Tool-Augmented LLMs
by: Cheng, Jiali, et al.
Published: (2025)

Split, Unlearn, Merge: Leveraging Data Attributes for More Effective Unlearning in LLMs
by: Kadhe, Swanand Ravindra, et al.
Published: (2024)

Steering LLMs for Formal Theorem Proving
by: Kirtania, Shashank, et al.
Published: (2025)

Measuring the Depth of LLM Unlearning via Activation Patching
by: Lee, Jaeung, et al.
Published: (2026)

SteeringSafety: A Systematic Safety Evaluation Framework of Representation Steering in LLMs
by: Siu, Vincent, et al.
Published: (2025)

How Data Inter-connectivity Shapes LLMs Unlearning: A Structural Unlearning Perspective
by: Qiu, Xinchi, et al.
Published: (2024)

Steering Llama 2 via Contrastive Activation Addition
by: Panickssery, Nina, et al.
Published: (2023)

Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs
by: Han, Pengrui, et al.
Published: (2026)

Focus On This, Not That! Steering LLMs with Adaptive Feature Specification
by: Lamb, Tom A., et al.
Published: (2024)

A Generative Approach to LLM Harmfulness Mitigation with Red Flag Tokens
by: Dobre, David, et al.
Published: (2025)

Improving Instruction-Following in Language Models through Activation Steering
by: Stolfo, Alessandro, et al.
Published: (2024)

Multi-property Steering of Large Language Models with Dynamic Activation Composition
by: Scalena, Daniel, et al.
Published: (2024)

Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering
by: Bigelow, Eric, et al.
Published: (2025)

Interpretable Steering of Large Language Models with Feature Guided Activation Additions
by: Soo, Samuel, et al.
Published: (2025)

ContextFocus: Activation Steering for Contextual Faithfulness in Large Language Models
by: Anand, Nikhil, et al.
Published: (2026)

SEUF: Is Unlearning One Expert Enough for Mixture-of-Experts LLMs?
by: Zhuang, Haomin, et al.
Published: (2024)

Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs
by: Siddique, Zara, et al.
Published: (2025)

The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence
by: Wollschläger, Tom, et al.
Published: (2025)

Extending Activation Steering to Broad Skills and Multiple Behaviours
by: van der Weij, Teun, et al.
Published: (2024)

Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMs
by: Xu, Xiaoyu, et al.
Published: (2025)

The Information Geometry of Softmax: Probing and Steering
by: Park, Kiho, et al.
Published: (2026)

Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space
by: Schwinn, Leo, et al.
Published: (2024)

UnStar: Unlearning with Self-Taught Anti-Sample Reasoning for LLMs
by: Sinha, Yash, et al.
Published: (2024)

Steering LLMs? Actually, Sparse Autoencoders can outperform simple baselines
by: Jørgensen, Mikkel Godsk, et al.
Published: (2026)

How to Train Your Advisor: Steering Black-Box LLMs with Advisor Models
by: Asawa, Parth, et al.
Published: (2025)

Learning to Steer: Input-dependent Steering for Multimodal LLMs
by: Parekh, Jayneel, et al.
Published: (2025)

Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
by: Karvonen, Adam, et al.
Published: (2025)

Towards Robust Evaluation of Unlearning in LLMs via Data Transformations
by: Joshi, Abhinav, et al.
Published: (2024)