English

On the Robustness Tradeoff in Fine-Tuning

Machine Learning 2025-07-15 v2 Computer Vision and Pattern Recognition

Abstract

Fine-tuning has become the standard practice for adapting pre-trained models to downstream tasks. However, the impact on model robustness is not well understood. In this work, we characterize the robustness-accuracy trade-off in fine-tuning. We evaluate the robustness and accuracy of fine-tuned models over 6 benchmark datasets and 7 different fine-tuning strategies. We observe a consistent trade-off between adversarial robustness and accuracy. Peripheral updates such as BitFit are more effective for simple tasks -- over 75% above the average measured by the area under the Pareto frontiers on CIFAR-10 and CIFAR-100. In contrast, fine-tuning information-heavy layers, such as attention layers via Compacter, achieves a better Pareto frontier on more complex tasks -- 57.5% and 34.6% above the average on Caltech-256 and CUB-200, respectively. Lastly, we observe that the robustness of fine-tuning against out-of-distribution data closely tracks accuracy. These insights emphasize the need for robustness-aware fine-tuning to ensure reliable real-world deployments.

Keywords

Cite

@article{arxiv.2503.14836,
  title  = {On the Robustness Tradeoff in Fine-Tuning},
  author = {Kunyang Li and Jean-Charles Noirot Ferrand and Ryan Sheatsley and Blaine Hoak and Yohan Beugin and Eric Pauley and Patrick McDaniel},
  journal= {arXiv preprint arXiv:2503.14836},
  year   = {2025}
}

Comments

Accepted to International Conference on Computer Vision, ICCV 2025

R2 v1 2026-06-28T22:26:08.336Z