TailLoR: Protecting Principal Components in Parameter-Efficient Continual Learning

Published
Source
arXiv
Paper number
325
Field
Machine Learning
arXiv ID
2606.06494

Key points

  • It designs low-rank updates by using the singular-value basis, U and V, of the pretrained weights as a fixed reference frame.
  • A soft spectral penalty suppresses updates along principal-component directions to reduce interference between tasks.
  • Fine-grained adaptation is routed into the long-tail spectral coordinates that have higher flexibility.
  • It improves continual-learning performance for PEFT methods based on spectral decomposition.

Paper links

External research summaries. These are not HDATF publications or measured product results.

Read original (opens in a new tab)