# DRIFT: Divergent Response in Filtered Transformations for Robust Adversarial Defense

arXiv: https://arxiv.org/abs/2509.24359 · HTML: https://mlsanity.com/paper/2509.24359/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Builds on (most cited first, 20 of 20)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks](https://mlsanity.com/paper/2003.01690/) (2020)
- [Certified Adversarial Robustness via Randomized Smoothing](https://mlsanity.com/paper/1902.02918/) (2019)
- [Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples](https://mlsanity.com/paper/1802.00420/) (2018)
- [Square Attack: a query-efficient black-box adversarial attack via random search](https://mlsanity.com/paper/1912.00049/) (2019)
- [On Adaptive Attacks to Adversarial Example Defenses](https://mlsanity.com/paper/2002.08347/) (2020)
- [Countering Adversarial Images using Input Transformations](https://mlsanity.com/paper/1711.00117/) (2017)
- [Diffusion Models for Adversarial Purification](https://mlsanity.com/paper/2205.07460/) (2022)
- [Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attack](https://mlsanity.com/paper/1907.02044/) (2019)
- [Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers](https://mlsanity.com/paper/1906.04584/) (2019)
- [Enhancing the Transferability of Adversarial Attacks through Variance Tuning](https://mlsanity.com/paper/2103.15571/) (2021)
- [A study of the effect of JPG compression on adversarial images](https://mlsanity.com/paper/1608.00853/) (2016)
- [Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNets](https://mlsanity.com/paper/2002.05990/) (2020)
- [DiffDefense: Defending against Adversarial Attacks via Diffusion Models](https://mlsanity.com/paper/2309.03702/) (2023)
- [Certifying Adapters: Enabling and Enhancing the Certification of Classifier Adversarial Robustness](https://mlsanity.com/paper/2405.16036/) (2024)
- [First line of defense: A robust first layer mitigates adversarial attacks](https://mlsanity.com/paper/2408.11680/) (2024)

## Built on by (most cited first, 1 of 1)

- [AEGIS: A Semantic GAN and Evidential Learning Frameworkfor Robust Adversarial Detection in Vision Sensors](https://mlsanity.com/paper/2606.28416/) (2026)

## Abstract (arXiv)

Deep neural networks remain highly vulnerable to adversarial examples, and most defenses collapse once gradients can be reliably estimated. We identify gradient consensus -- the tendency of randomized transformations to yield aligned gradients -- as a key driver of adversarial transferability. Attackers exploit this consensus to construct perturbations that remain effective across transformations. We introduce DRIFT (Divergent Response in Filtered Transformations), a stochastic ensemble of lightweight, learnable filters trained to actively disrupt gradient consensus. Unlike prior randomized defenses that rely on gradient masking, DRIFT enforces gradient dissonance by maximizing divergence in Jacobian- and logit-space responses while preserving natural predictions. Our contributions are threefold: (i) we formalize gradient consensus and provide a theoretical analysis linking consensus to transferability; (ii) we propose a consensus-divergence training strategy combining prediction consistency, Jacobian separation, logit-space separation, and adversarial robustness; and (iii) we show that DRIFT achieves substantial robustness gains on ImageNet across CNNs and Vision Transformers, outperforming state-of-the-art preprocessing, adversarial training, and diffusion-based defenses under adaptive white-box, transfer-based, and gradient-free attacks. DRIFT delivers these improvements with negligible runtime and memory cost, establishing gradient divergence as a practical and generalizable principle for adversarial defense.