mlsanityData to 14 Oct 2026

Adaptive Diffusion Denoised Smoothing : Certified Robustness via Randomized Smoothing with Differentially Private Guided Denoising Diffusion

arXiv 2507.08163 ↗First posted 10 Jul 2025cs.CV cs.CR cs.LGcited by 0

On the map: Diffusion models → Robustness, privacy and adversarial aspects

Abstract (arXiv)

We propose Adaptive Diffusion Denoised Smoothing, a method for certifying the predictions of a vision model against adversarial examples, while adapting to the input. Our key insight is to reinterpret a guided denoising diffusion model as a long sequence of adaptive Gaussian Differentially Private (GDP) mechanisms refining a pure noise sample into an image. We show that these adaptive mechanisms can be composed through a GDP privacy filter to analyze the end-to-end robustness of the guided denoising process, yielding a provable certification that extends the adaptive randomized smoothing analysis. We demonstrate that our design, under a specific guiding strategy, can improve both certified accuracy and standard accuracy on ImageNet for an ℓ2 threat model.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 22 of its 27 references that are in the map. First posted on arXiv 10 Jul 2025.

Builds on

22 of 27 references.

Closest ancestors · 5

Same direction, newest first

  1. 2023Multi-scale Diffusion Denoised Smoothing25 Oct
  2. 2023DiffSmooth: Certifiably Robust Learning via Diffusion Models and Local Smoothing28 Aug
  3. 2022Guided Diffusion Model for Adversarial Purification from Random Noise22 Jun
  4. 2022(Certified!!) Adversarial Robustness for Free!21 Jun
  5. 2022Guided Diffusion Model for Adversarial Purification30 May

Other references · 17

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Denoising Diffusion Implicit Modelscited by 9,781
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2013Intriguing properties of neural networkscited by 5,155
  5. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  6. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,375
  7. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  8. 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 468
  9. 2023Universal Guidance for Diffusion Modelscited by 437
  10. 2023Are aligned neural networks adversarially aligned?cited by 315
  11. 2022Evaluating the Adversarial Robustness of Adaptive Test-time Defensescited by 64
  12. 2020Data-Dependent Randomized Smoothingcited by 37
5 more
  1. 2016Privacy Odometers and Filters: Pay-as-you-Go Compositioncited by 33
  2. 2022Individual Privacy Accounting with Gaussian Differential Privacycited by 20
  3. 2024Adaptive Randomized Smoothing: Certified Adversarial Robustness for Multi-Step Defencescited by 9
  4. 2022Fully Adaptive Composition for Gaussian Differential Privacycited by 5
  5. 2023Concurrent Composition for Interactive Differential Privacy with Adaptive Privacy-Loss Parameterscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Beyond Classification: Evaluating Diffusion Denoised Smoothing for Security-Utility Trade off
  2. 2025Robustifying Diffusion-Denoised Smoothing Against Covariate Shift
  3. 2024Gradient-Guided Conditional Diffusion Models for Private Image Reconstruction: Analyzing Adversarial Impacts of Differential Privacy and Denoising
  4. 2025Robust Representation Consistency Model via Contrastive Denoising
  5. 2025Demystifying Adversarial Robustness in Diffusion Models: Compression, Randomness, and Geometry
  6. 2022DensePure: Understanding Diffusion Models towards Adversarial Robustness
  7. 2024Adversarial Robustification via Text-to-Image Diffusion Models
  8. 2024Differentially Private Fine-Tuning of Diffusion Models