mlsanityData to 14 Oct 2026

Enhancing Adversarial Training with Prior Knowledge Distillation for Robust Image Compression

arXiv 2403.06700 ↗First posted 11 Mar 2024eess.IVcited by 1

On the map: Knowledge distillation → Federated, robust and private distillation

Abstract (arXiv)

Deep neural network-based image compression (NIC) has achieved excellent performance, but NIC method models have been shown to be susceptible to backdoor attacks. Adversarial training has been validated in image compression models as a common method to enhance model robustness. However, the improvement effect of adversarial training on model robustness is limited. In this paper, we propose a prior knowledge-guided adversarial training framework for image compression models. Specifically, first, we propose a gradient regularization constraint for training robust teacher models. Subsequently, we design a knowledge distillation based strategy to generate a priori knowledge from the teacher model to the student model for guiding adversarial training. Experimental results show that our method improves the reconstruction quality by about 9dB when the Kodak dataset is elected as the backdoor attack object for psnr attack. Compared with Ma2023, our method has a 5dB higher PSNR output at high bitrate points.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Federated, robust and private distillation, one of 12 directions in Knowledge distillation. Builds on 12 of its 20 references that are in the map. First posted on arXiv 11 Mar 2024.

Builds on

12 of 20 references.

References · 12

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2013Intriguing properties of neural networkscited by 5,155
  4. 2016Adversarial examples in the physical worldcited by 1,769
  5. 2018Variational image compression with a scale hyperpriorcited by 935
  6. 2016End-to-end Optimized Image Compressioncited by 696
  7. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 678
  8. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 664
  9. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 590
  10. 2020Channel-wise Autoregressive Entropy Models for Learned Image Compressioncited by 264
  11. 2019Scaleable input gradient regularization for adversarial robustnesscited by 47
  12. 2021Towards Robust Neural Image Compression: Adversarial Attack and Model Finetuningcited by 25

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Fast Test-Time Refinement for Robust Learned Image Compressioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Microdosing: Knowledge Distillation for GAN based Compression
  2. 2025Efficient Learned Image Compression Through Knowledge Distillation
  3. 2021How and When Adversarial Robustness Transfers in Knowledge Distillation?
  4. 2022On the benefits of knowledge distillation for adversarial robustness
  5. 2021Countering Adversarial Examples: Combining Input Transformation and Noisy Training
  6. 2025Improving Adversarial Robustness Through Adaptive Learning-Driven Multi-Teacher Knowledge Distillation
  7. 2023Mitigating Accuracy-Robustness Trade-off via Balanced Multi-Teacher Adversarial Distillation
  8. 2026Learning Better Certified Models from Empirically-Robust Teachers