Activation engineering enables precise control over Large Language Models (LLMs) without the computational cost of fine-tuning. However, existing methods deriving vectors from static activation differences are susceptible to high-dimensional noise and layer-wise semantic drift, often capturing spurious correlations rather than the target intent. To address this, we propose Global Evolutionary Refined Steering (GER-steer), a training-free framework that grounded in the geometric stability of the network's representation evolution. GER-steer exploits this global signal to rectify raw steering vectors, effectively decoupling robust semantic intent from orthogonal artifacts. Extensive evaluations confirm that GER-steer consistently outperforms baselines, delivering superior efficacy and generalization without layer-specific tuning, establishing a universal solution for reliable model alignment.
Cite
@article{arxiv.2603.12298,
title = {Global Evolutionary Steering: Refining Activation Steering Control via Cross-Layer Consistency},
author = {Xinyan Jiang and Wenjing Yu and Di Wang and Lijie Hu},
journal= {arXiv preprint arXiv:2603.12298},
year = {2026}
}