Image de couverture de livre

Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Bao, Yuntai, Li, Qinfeng, Yu, Xinyan, Zhang, Xuhong, Su, Ge, Zhang, Wenqi, Yan, Liu, Weng, Haiqin, Yin, Jianwei
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Faithful Bi-Directional Model Steering via Distribution Matching and Distributed Interchange Interventions
par: Bao, Yuntai, et autres
Publié: (2026)

PragLocker: Protecting Agent Intellectual Property in Untrusted Deployments via Non-Portable Prompts
par: Li, Qinfeng, et autres
Publié: (2026)

Beyond Steering Vector: Flow-based Activation Steering for Inference-Time Intervention
par: Jin, Zehao, et autres
Publié: (2026)

Scalable Multi-Stage Influence Function for Large Language Models via Eigenvalue-Corrected Kronecker-Factored Parameterization
par: Bao, Yuntai, et autres
Publié: (2025)

Probing the Geometry of Truth: Consistency and Generalization of Truth Directions in LLMs Across Logical Transformations and Question Answering Tasks
par: Bao, Yuntai, et autres
Publié: (2025)

Steer2Edit: From Activation Steering to Component-Level Editing
par: Sun, Chung-En, et autres
Publié: (2026)

Global Evolutionary Steering: Refining Activation Steering Control via Cross-Layer Consistency
par: Jiang, Xinyan, et autres
Publié: (2026)

EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering
par: Xu, Haolei, et autres
Publié: (2025)

Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization
par: Cao, Yuanpu, et autres
Publié: (2024)

AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint
par: Sheng, Leheng, et autres
Publié: (2025)

Steer Like the LLM: Activation Steering that Mimics Prompting
par: Heyman, Geert, et autres
Publié: (2026)

Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectors
par: Wang, Weixuan, et autres
Publié: (2024)

Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs
par: Han, Pengrui, et autres
Publié: (2026)

Towards Reliable Evaluation of Behavior Steering Interventions in LLMs
par: Pres, Itamar, et autres
Publié: (2024)

Persona Vectors in Games: Measuring and Steering Strategies via Activation Vectors
par: Sun, Johnathan, et autres
Publié: (2026)

No Training Wheels: Steering Vectors for Bias Correction at Inference Time
par: Gupta, Aviral, et autres
Publié: (2025)

HijackRAG: Hijacking Attacks against Retrieval-Augmented Large Language Models
par: Zhang, Yucheng, et autres
Publié: (2024)

VSPO: Vector-Steered Policy Optimization for Behavioral Control
par: Zhang, Xuechen, et autres
Publié: (2026)

SteerX: Disentangled Steering for LLM Personalization
par: Zhao, Xiaoyan, et autres
Publié: (2025)

TransLinkGuard: Safeguarding Transformer Models Against Model Stealing in Edge Deployment
par: Li, Qinfeng, et autres
Publié: (2024)

FineSteer: A Unified Framework for Fine-Grained Inference-Time Steering in Large Language Models
par: Weng, Zixuan, et autres
Publié: (2026)

Subliminal Learning Is Steering Vector Distillation
par: Blank, Camila, et autres
Publié: (2026)

Analysing the Safety Pitfalls of Steering Vectors
par: Li, Yuxiao, et autres
Publié: (2026)

Analyzing the Generalization and Reliability of Steering Vectors
par: Tan, Daniel, et autres
Publié: (2024)

Predicting Where Steering Vectors Succeed
par: Billa, Jayadev
Publié: (2026)

Prompt-Activation Duality: Improving Activation Steering via Attention-Level Interventions
par: Kang, Diancheng, et autres
Publié: (2026)

Steering-induced phase transition in measurement-only quantum circuits
par: Qian, Dongheng, et autres
Publié: (2023)

FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering
par: Li, Yichen, et autres
Publié: (2025)

Spectral Attention Steering for Prompt Highlighting
par: Li, Weixian Waylon, et autres
Publié: (2026)

White-Box Sensitivity Auditing with Steering Vectors
par: Cyberey, Hannah, et autres
Publié: (2026)

Dialz: A Python Toolkit for Steering Vectors
par: Siddique, Zara, et autres
Publié: (2025)

CogSteer: Cognition-Inspired Selective Layer Intervention for Efficiently Steering Large Language Models
par: Wang, Xintong, et autres
Publié: (2024)

PrimeComposer: Faster Progressively Combined Diffusion for Image Composition with Attention Steering
par: Wang, Yibin, et autres
Publié: (2024)

Neural Quantum States in Variational Monte Carlo Method: A Brief Summary
par: Song, Yuntai
Publié: (2024)

Adaptive Multi-Subspace Representation Steering for Attribute Alignment in Large Language Models
par: Jiang, Xinyan, et autres
Publié: (2025)

MedSteer: Counterfactual Endoscopic Synthesis via Training-Free Activation Steering
par: Pham, Trong-Thang, et autres
Publié: (2026)

Where to Steer: Input-Dependent Layer Selection for Steering Improves LLM Alignment
par: Gadgil, Soham, et autres
Publié: (2026)

ROAST: Rollout-based On-distribution Activation Steering Technique
par: Su, Xuanbo, et autres
Publié: (2026)

Towards Understanding Steering Strength
par: Taimeskhanov, Magamed, et autres
Publié: (2026)

Causal Interventions on Continuous Variables: A Case Study on Verb Bias in Steering Vectors for In-Context Learning
par: Zhou, Zhenghao Herbert, et autres
Publié: (2026)