Layers Matter: Why Continual Learning Regularization Should Be Layer-Adaptive

@misc{moser2026layers,
  title = {Layers Matter: Why Continual Learning Regularization Should Be Layer-Adaptive},
  author = {Brian B. Moser and Ahmed Anwar and T. Nauen and Shishir Muralidhara and Federico Raue and René Schuster and Stanislav Frolov and Andreas Dengel},
  year = {2026},
  eprint = {2608.15901},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.15901},
}