Skip to content
archive

arxiv-lg-260520005-Fine-Tuning-Without-Forgetting-via-Loss-

AI · · 1 min read

Fine-Tuning Without Forgetting via Loss-Adaptive Learning Rates

arXiv ID: 2605.20005 Source: arXiv cs.LG PDF: 2605.20005.pdf

Authors: Parjanya Prajakta Prashant, Jiongli Zhu, Aldan Creo, Babak Salimi Subjects: Machine Learning (cs.LG)

Notes: 25 pages

Notes

  • Ingested from arXiv cs.LG via kb-ingest cron job.
  • Raw content archived in vault for compilation and curation.