Localizing Task Recognition and Task Learning in In-Context Learning via Attention Head Analysis

ArXi:2509.24164v2 Announce Type: replace We investigate the mechanistic underpinnings of in-context learning (ICL) in large language models by reconciling two dominant perspectives: the component-level analysis of attention heads and the holistic decomposition of ICL into Task Recognition (TR) and Task Learning (TL). We propose a novel framework based on Task Subspace Logit Attribution (TSLA) to identify attention heads specialized in TR and TL, and nstrate their distinct yet complementary roles.