LLMs have shown remarkable capabilities, but precisely controlling their response behavior remains challenging. Existing activation steering methods alter LLM behavior indiscriminately, limiting their practical applicability in settings where selective responses are essential, such as content moderation or domain-specific assistants. In this paper, we propose Conditional Activation Steering (CAST), which analyzes LLM activation patterns during inference to selectively apply or withhold activation steering based on the input context. Our method is based on the observation that different categories of prompts activate distinct patterns in the model's hidden states. Using CAST, one can systematically control LLM behavior with rules like "if input is about hate speech or adult content, then refuse" or "if input is not about legal advice, then refuse." This allows for selective modification of responses to specific content while maintaining normal responses to other content, all without requiring weight optimization. We release an open-source implementation of our framework at github.com/IBM/activation-steering .
Graph-Regularized Sparse Autoencoders for LLM Safety Steering
ELLIS Institute Tübingen（图宾根ELLIS研究所）；Max Planck Institute for Intelligent Systems（智能系统马克斯·普朗克研究所）；Intesa Sanpaolo（Intesa Sanpaolo公司）；University of Southern California（南加州大学）
Interpretable Steering of Large Language Models with Feature Guided Activation Additions
Raffles Science Institute（莱佛士科学学院）；Raffles Institution（莱佛士书院）；Nous Research（努斯研究机构）；Centre for Frontier AI Research (CFAR)（前沿人工智能研究中心）；Agency for Science Technology and Research (A*STAR)（新加坡科技研究局）
Torilab；Department of Mathematics National University of Singapore（数学系，新加坡国立大学）
Controlling specific behaviors in large language models while preserving their general capabilities is a central challenge for safe and reliable artificial intelligence deployment. Current steering methods, such as vector addition and directional ablation, are constrained within a two-dimensional subspace defined by the activation and feature direction, making them sensitive to chosen parameters and potentially affecting unrelated features due to unintended interactions in activation space. We introduce Angular Steering, a novel and flexible method for behavior modulation that operates by rota…
EmoRSS: Mitigating Emotion-Induced Over-Refusal in Large Language Models
Beihang University（北京航空航天大学）；University of Science and Technology Beijing（北京科技大学）；National University of Singapore（新加坡国立大学）；Lanzhou University（兰州大学）；Jinan University（暨南大学）
Activation Steering Induces Emergent Misalignment: Steering Vector Construction and Comprehensive Evaluation
Nanyang Technological University（南洋理工大学）；Sun Yat-sen University（中山大学）；University of Science and Technology of China（中国科学技术大学）；National University of Singapore（新加坡国立大学）
Minimizing Collateral Damage in Activation Steering
Department of Electrical \& Computer Engineering, Rice University, Houston, USA；Department of Computational；Applied Mathematics, Rice University, Houston, USA；Department of Electrical \& Computer Engineering, The University of Texas at Austin, Austin, USA
ODESteer: A Unified ODE-Based Steering Framework for LLM Alignment
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Northwestern University（西北大学）；William & Mary（威廉与玛丽学院）；University of Washington（华盛顿大学）；Stanford University（斯坦福大学）
Nullify: Null-Space Activation Steering for Training-Free LLM Unlearning
College of Computer Science and Artificial Intelligence, Fudan University（复旦大学计算机与人工智能学院）；BEDI Cloud（BEDI云公司）；School of Electrical and Information Engineering, Tianjin University（天津大学电气与信息工程学院）；King Abdullah University of Science and Technology (KAUST)（阿卜杜拉国王科技大学）；School of Software Technology, Zhejiang University（浙江大学软件学院）；School of Transnational Law, Peking University（北京大学国际法学院）
