
Selectively regularizing parameter updates during fine-tuning, as opposed to applying uniform regularization, leads to improved in-distribution generalization and out-of-distribution robustness in foundation models.


coremsg

selective-projection-decay-a-novel-weight-decay-technique-for-robust-fine-tuning-of-foundation-models


Selective Projection Decay: A Novel Weight Decay Technique for Robust Fine-Tuning of Foundation Models


title_rewrite


AUTOFT is a data-driven approach that significantly improves generalization to out-of-distribution inputs, surpassing existing methods.


autoft-learning-an-objective-for-robust-fine-tuning


AutoFT: Learning an Objective for Robust Fine-Tuning
