AI RESEARCH
The Cylindrical Representation Hypothesis for Language Model Steering
arXiv CS.CL
•
ArXi:2605.01844v1 Announce Type: new Steering is a widely used technique for controlling large language models, yet its effects are often unstable and hard to predict. Existing theoretical accounts are largely based on the Linear Representation Hypothesis (LRH). While LRH assumes that concepts can be orthogonalized for lossless control, this idealized mapping fails in real representations and cannot account for the observed unpredictability of steering.