activation steering

SemEval-2026 Task 11: Disentangling Content and Formal Reasoning in Large Language Models SemEval 2026 Silenced Biases: The Dark Side LLMs Learned to Refuse AAAI 2026 SteerForce at SemEval-2026 Task 11: Reducing Content Effects Using Layered Activation Steering SemEval 2026 Test-time Diverse Reasoning by Riemannian Activation Steering AAAI 2026 Activation Steering Decoding: Mitigating Hallucination in Large Vision-Language Models through Bidirectional Hidden State Intervention ACL 2025 Activation Steering Meets Preference Optimization: Defense Against Jailbreaks in Vision Language Model ACML 2025 AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender EMNLP 2025 A Simple Yet Effective Method for Non-Refusing Context Relevant Fine-grained Safety Steering in LLMs EMNLP 2025 Breaking Bad Tokens: Detoxification of LLMs Using Sparse Autoencoders EMNLP 2025 Can Role Vectors Affect LLM Behaviour? EMNLP 2025 Controlling Language and Diffusion Models by Transporting Activations ICLR 2025 COSMIC: Generalized Refusal Direction Identification in LLM Activations ACL 2025 Distributional Surgery for Language Model Activations EMNLP 2025 Dynamic Steering With Episodic Memory For Large Language Models ACL 2025 FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering ACL 2025 Improving Instruction-Following in Language Models through Activation Steering ICLR 2025 Internal Chain-of-Thought: Empirical Evidence for Layer‐wise Subtask Scheduling in LLMs EMNLP 2025 Internal Value Alignment in Large Language Models through Controlled Value Vector Activation ACL 2025 Mechanistic Interpretability for Steering Vision-Language-Action Models CoRL 2025 Personalized Text Generation with Contrastive Activation Steering ACL 2025 Pixels Versus Priors: Controlling Knowledge Priors in Vision-Language Models through Visual Counterfacts EMNLP 2025 Programming Refusal with Conditional Activation Steering ICLR 2025 Representation Bending for Large Language Model Safety ACL 2025 SAKE: Steering Activations for Knowledge Editing ACL 2025 SCANS: Mitigating the Exaggerated Safety for LLMs via Safety-Conscious Activation Steering AAAI 2025