mlsanityData to 7 Oct 2026

A Survey on Gradient Inversion: Attacks, Defenses and Future Directions

arXiv 2206.07284 ↗First posted 15 Jun 2022cs.LGcited by 39

Not on the map. Why?

Abstract (arXiv)

Recent studies have shown that the training samples can be recovered from gradients, which are called Gradient Inversion (GradInv) attacks. However, there remains a lack of extensive surveys covering recent advances and thorough analysis of this issue. In this paper, we present a comprehensive survey on GradInv, aiming to summarize the cutting-edge research and broaden the horizons for different domains. Firstly, we propose a taxonomy of GradInv attacks by characterizing existing attacks into two paradigms: iteration- and recursion-based attacks. In particular, we dig out some critical ingredients from the iteration-based attacks, including data initialization, model training and gradient matching. Second, we summarize emerging defense strategies against GradInv attacks. We find these approaches focus on three perspectives covering data obscuration, model improvement and gradient protection. Finally, we discuss some promising directions and open problems for further research.

Key facts. Cited by 39 papers in the map (data through 7 Oct 2026). Builds on 39 of its 59 references that are in the map. First posted on arXiv 15 Jun 2022.

Builds on

39 of 59 references.

References · 39

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  5. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  6. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  7. 2019Deep Leakage from Gradientscited by 1,004
  8. 2019Towards Federated Learning at Scale: System Designcited by 826
  9. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
  10. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  11. 2019Federated Learning with Differential Privacy: Algorithms and Performance Analysiscited by 528
  12. 2014Understanding Deep Image Representations by Inverting Themcited by 399
27 more
  1. 2020iDLG: Improved Deep Leakage from Gradientscited by 367
  2. 2021See through Gradients: Image Batch Recovery via GradInversioncited by 320
  3. 2019Error Feedback Fixes SignSGD and other Gradient Compression Schemescited by 306
  4. 2019PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimizationcited by 234
  5. 2021Adversary Instantiation: Lower Bounds for Differentially Private Machine Learningcited by 193
  6. 2020LDP-Fed: Federated Learning with Local Differential Privacycited by 129
  7. 2020R-GAP: Recursive Gradient Attack on Privacycited by 95
  8. 2020Provable Defense against Privacy Leakage in Federated Learning from Representation Perspectivecited by 94
  9. 2020InstaHide: Instance-hiding Schemes for Private Distributed Learningcited by 90
  10. 2021TAG: Gradient Attack on Transformer-based Language Modelscited by 63
  11. 2019Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attackscited by 54
  12. 2021Survey: Leakage and Privacy at Inference Timecited by 44
  13. 2021Gradient Disaggregation: Breaking Privacy in Federated Learning by Reconstructing the User Participant Matrixcited by 43
  14. 2021Gradient-Leakage Resilient Federated Learningcited by 40
  15. 2021Bayesian Framework for Gradient Leakagecited by 34
  16. 2020Rethinking Privacy Preserving Deep Learning: How to Evaluate and Thwart Privacy Attackscited by 32
  17. 2020SAPAG: A Self-Adaptive Privacy Attack From Gradientscited by 27
  18. 2021APRIL: Finding the Achilles' Heel on Privacy for Vision Transformerscited by 23
  19. 2020Layer-wise Characterization of Latent Information Leakage in Federated Learningcited by 19
  20. 2021Revealing and Protecting Labels in Distributed Trainingcited by 17
  21. 2021Understanding Training-Data Leakage from Gradients in Neural Networks for Image Classificationcited by 16
  22. 2020Tighter Generalization Bounds for Iterative Differentially Private Learning Algorithmscited by 14
  23. 2021H-FL: A Hierarchical Communication-Efficient and Privacy-Protected Architecture for Federated Learningcited by 14
  24. 2021A Quantitative Metric for Privacy Leakage in Federated Learningcited by 12
  25. 2020Minimal Model Structure Analysis for Input Reconstruction in Federated Learningcited by 5
  26. 2021Federated Learning Attacks Revisited: A Critical Discussion of Gaps, Assumptions, and Evaluation Setupscited by 5
  27. 2021Dropout against Deep Leakage from Gradientscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 11; 2024: 13; 2025: 8; 2026: 3.

39 papers.

Sort:

By area:

  1. 2023Dataset Distillation: A Comprehensive Reviewcited by 126
  2. 2023GIFD: A Generative Gradient Inversion Method with Feature Domain Optimizationcited by 45
  3. 2023Privacy Assessment on Reconstructed Images: Are Existing Evaluation Metrics Faithful to Human Perception?cited by 13
  4. 2024GI-NAS: Boosting Gradient Inversion Attacks Through Adaptive Neural Architecture Searchcited by 13
  5. 2022Dropout is NOT All You Need to Prevent Gradient Leakagecited by 11
  6. 2023The Resource Problem of Using Linear Layer Leakage Attack in Federated Learningcited by 9
  7. 2023A Survey on Vulnerability of Federated Learning: A Learning Algorithm Perspectivecited by 9
  8. 2025Exploring the Vulnerabilities of Federated Learning: A Deep Dive into Gradient Inversion Attackscited by 9
  9. 2022How Does a Deep Learning Model Architecture Impact Its Privacy? A Comprehensive Study of Privacy Attacks on CNNs and Transformerscited by 8
  10. 2025RAG Security and Privacy: Formalizing the Threat Model and Attack Surfacecited by 8
  11. 2023Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systemscited by 6
  12. 2023Client-side Gradient Inversion Against Federated Learning from Poisoningcited by 4
27 more
  1. 2023Transpose Attack: Stealing Datasets with Bidirectional Trainingcited by 4
  2. 2024Towards Eliminating Hard Label Constraints in Gradient Inversion Attackscited by 4
  3. 2023On Knowledge Editing in Federated Learning: Perspectives, Challenges, and Future Directionscited by 3
  4. 2024Memory Backdoor Attacks on Neural Networkscited by 3
  5. 2024Geminio: Language-Guided Gradient Inversion Attacks in Federated Learningcited by 3
  6. 2022Combining Stochastic Defenses to Resist Gradient Inversion: An Ablation Studycited by 2
  7. 2022Refiner: Data Refining against Gradient Leakage Attacks in Federated Learningcited by 2
  8. 2023Privacy Preserving Federated Learning with Convolutional Variational Bottleneckscited by 2
  9. 2024AFGI: Towards Accurate and Fast-convergent Gradient Inversion Attack in Federated Learningcited by 2
  10. 2024GI-SMN: Gradient Inversion Attack against Federated Learning without Prior Knowledgecited by 2
  11. 2024Dealing Doubt: Unveiling Threat Models in Gradient Inversion Attacks under Federated Learning, A Survey and Taxonomycited by 2
  12. 2025Adversarial Robustness in Distributed Quantum Machine Learningcited by 2
  13. 2025The MICCAI Federated Tumor Segmentation (FeTS) Challenge 2024: Efficient and Robust Aggregation Methods for Federated Learningcited by 2
  14. 2026Enhancing Gradient Inversion Attacks in Federated Learning via Hierarchical Feature Optimizationcited by 2
  15. 2023RAIFLE: Reconstruction Attacks on Interaction-based Federated Learning with Adversarial Data Manipulationcited by 1
  16. 2024LDPKiT: Superimposing Remote Queries for Privacy-Preserving Distillationcited by 1
  17. 2024Intermediate Outputs Are More Sensitive Than You Thinkcited by 1
  18. 2024FedGIG: Graph Inversion from Gradient in Federated Learningcited by 1
  19. 2026Reconstructing Training Data from Adapter-based Federated Large Language Modelscited by 1
  20. 2024Optimally Improving Cooperative Learning in a Social Settingcited by 0
  21. 2024Optimal Defenses Against Gradient Reconstruction Attackscited by 0
  22. 2024GDBR: Label Recovery Attack Against Partial Gradient Encryption in Federated Learningcited by 0
  23. 2025PC-MoE: Memory-Efficient and Privacy-Preserving Collaborative Training for Mixture-of-Experts LLMscited by 0
  24. 2025Evaluating Selective Encryption Against Gradient Inversion Attackscited by 0
  25. 2025Images in Motion?: A First Look into Video Leakage in Collaborative Deep Learningcited by 0
  26. 2025GraphToxin: Reconstructing Full Unlearned Graphs from Graph Unlearningcited by 0
  27. 2026Leaking Circuit Secrets: Gradient Leakage Attacks on Graph Neural Networkscited by 0