English

Formalising the Logit Shift Induced by LoRA: A Technical Note

Machine Learning 2026-04-23 v1 Artificial Intelligence

Abstract

This technical note provides a first-order formalisation of the logit shift and fact-margin change induced by Low-Rank Adaptation (LoRA). Using a first-order Fr\'echet approximation around the base model trajectory, we show that the multi-layer LoRA effect can be decomposed into a linear summation of layerwise contributions and a higher-order remainder term representing inter-layer coupling.

Cite

@article{arxiv.2604.20313,
  title  = {Formalising the Logit Shift Induced by LoRA: A Technical Note},
  author = {Xiang Shi and Shuaizhi Cheng and Mingwei Li},
  journal= {arXiv preprint arXiv:2604.20313},
  year   = {2026}
}

Comments

7 pages, technical note

R2 v1 2026-07-01T12:29:57.974Z