mlsanityData to 7 Oct 2026

On the Robustness Tradeoff in Fine-Tuning

arXiv 2503.14836 ↗First posted 19 Mar 2025cs.LG cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 36 of its 51 references that are in the map. First posted on arXiv 19 Mar 2025.

Builds on

36 of 51 references.

References · 36

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  7. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  8. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  9. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  10. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  11. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  12. 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,327
24 more
  1. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  2. 2022Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learningcited by 886
  3. 2018Robustness May Be at Odds with Accuracycited by 873
  4. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  5. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 722
  6. 2021ImageNet-21K Pretraining for the Massescited by 607
  7. 2021Compacter: Efficient Low-Rank Hypercomplex Adapter Layerscited by 458
  8. 2019A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmarkcited by 419
  9. 2020Parameter-Efficient Transfer Learning with Diff Pruningcited by 351
  10. 2018Do CIFAR-10 Classifiers Generalize to CIFAR-10?cited by 197
  11. 2023Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuningcited by 171
  12. 2020Torchattacks: A PyTorch Repository for Adversarial Attackscited by 156
  13. 2020Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuningcited by 152
  14. 2020Robust Pre-Training by Adversarial Contrastive Learningcited by 144
  15. 2022KronA: Parameter Efficient Tuning with Kronecker Adaptercited by 130
  16. 2022Parameter-efficient Model Adaptation for Vision Transformerscited by 81
  17. 2014Fastfood: Approximate Kernel Expansions in Loglinear Timecited by 72
  18. 2020A Simple Fine-tuning Is All You Need: Towards Robust Deep Learning Via Adversarial Fine-tuningcited by 30
  19. 2023TWINS: A Fine-Tuning Framework for Improved Transferability of Adversarial Robustness and Generalizationcited by 18
  20. 2022To the Fairness Frontier and Beyond: Identifying, Quantifying, and Optimizing the Fairness-Accuracy Pareto Frontiercited by 10
  21. 2023Initialization Matters for Adversarial Transfer Learningcited by 9
  22. 2021CARTL: Cooperative Adversarially-Robust Transfer Learningcited by 8
  23. 2024Adapters Mixup: Mixing Parameter-Efficient Adapters to Enhance the Adversarial Robustness of Fine-tuned Pre-trained Text Classifierscited by 3
  24. 2024Criticality Leveraged Adversarial Training (CLAT) for Boosted Performance via Parameter Efficiencycited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Alignment midtraining for animalscited by 4
  2. 2026The Evaluation Game: Beyond Static LLM Benchmarkingcited by 0
  3. 2026Point-Selection Fine-Tuning Framework for Robust Point Cloud Classificationcited by 0