基本信息
- 来源: arxiv
- 原始来源: https://arxiv.org/abs/2606.06494v1
- 作者: Marius Dragoi, Ioana Pintilie, Alexandra Dragomir, Antonio Barbalau, Florin Brad
- 分类: cs.LG
- 论文时间: 2026-06-04T17:59:55Z
- 论文 PDF: https://arxiv.org/pdf/2606.06494v1.pdf
来源摘要/节选
Parameter-efficient finetuning methods based on spectral decomposition have enabled progress in Continual Learning. In this paper we introduce TailLoR, which utilizes the singular bases U and V of the pre-trained weights as a fixed reference frame to learn a low-rank update applied to the singular value matrix. A soft spectral penalty discourages updates aligned with dominant singular directions, reducing interference while routing fine-grained adaptation into the highly flexible, long-tail spectral coordinates.
来源说明
当前只保存了官方论文摘要，不代表论文全文。请以原始来源为准。
本页只呈现已做哈希绑定的来源证据，不包含基于旧正文或缺失原文的扩展推断。