AI RESEARCH
In-Context Learning Operates as Concept Subspace Learning
arXiv CS.LG
•
ArXi:2605.18830v1 Announce Type: new Regression and Bayesian accounts of in-context learning (ICL) explain how nstrations can induce predictors, while mechanistic analyses often identify compact activation directions that steer prompted behavior. However, it remains unclear whether structured nstrations induce low-dimensional concept inference. We study this question through a concept-subspace view of ICL, in which tasks vary only along intrinsic concept coordinates, although inputs are observed in a high-dimensional ambient space.