[2509.25175] EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering

Computer Science > Computation and Language arXiv:2509.25175 (cs) [Submitted on 29 Sep 2025 (v1), last revised 2 Mar 2026 (this version, v2)] Title:EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering Authors:Haolei Xu, Xinyu Mei, Yuchen Yan, Rui Zhou, Wenqi Zhang, Weiming Lu, Yueting Zhuang, Yongliang Shen View a PDF of the paper titled EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering, by Haolei Xu and 7 other authors View PDF HTML (experimental) Abstract:Large language model (LLM) steering has emerged as a promising paradigm for controlling model behavior at inference time through targeted manipulation of hidden states, offering a lightweight alternative to expensive retraining. However, existing steering frameworks suffer from critical limitations: computational inefficiency, limited extensibility, and restricted functionality that hinder both research progress and practical deployment. We present EasySteer, a unified framework for high-performance, extensible LLM steering built on vLLM. Our system features modular architecture with pluggable interfaces for both analysis-based and learning-based methods, fine-grained parameter control, pre-computed steering vectors for eight application domains, and an interactive demonstration system. Through deep integration with vLLM's optimized inference engine, EasySteer achieves 10.8-22.3$\times$ speedup over existing frameworks. Extensive experiments demonstrate its effect...