Lune

Abstract

Federated Learning (FL) is a promising paradigm for collaboratively training Machine Learning (ML) models while preserving the privacy of data owners. By allowing participants to maintain their data on-site, FL avoids sending client local data to a central server for model training. However, despite its evident privacy benefits, it is not immune to security and privacy threats. Among these, Gradient Inversion Attacks (GIAs) stand out as one of the most critical as they exploit client's model updates to reconstruct local training data, breaking participant's privacy. This work presents a comprehensive systematization of GIAs in FL. First, we identify various threat models defining the adversary's knowledge and capabilities to perform these attacks. Then, we propose a systematic taxonomy to categorize GIAs, providing practical insights into their methods and applicability. Additionally, we explore defensive mechanisms designed to mitigate these attacks. We also systematize evaluation metrics used to measure the success of GIAs and assess the model's vulnerability before an attack. Finally, based on a thorough analysis of the existing literature, we identify key challenges and outline promising future research directions.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

4 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning cites Practical Secure Aggregation for Privacy-Preserving Machine Learning
  • Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning cites Deep Learning with Differential Privacy
  • Practical Secure Aggregation for Privacy-Preserving Machine Learning cites Deep Learning with Differential Privacy
  • When Topology Betrays Privacy: Lattice-Based Reconstruction Attacks on Secure Aggregation in Decentralized Federated Learning cites Practical Secure Aggregation for Privacy-Preserving Machine Learning
  • When Topology Betrays Privacy: Lattice-Based Reconstruction Attacks on Secure Aggregation in Decentralized Federated Learning cites Inverting Gradients - How easy is it to break privacy in federated learning?
  • SCAFFOLD: Stochastic Controlled Averaging for Federated Learning cites Practical Secure Aggregation for Privacy-Preserving Machine Learning
  • Cracking Federated Privacy: Initialization-Resilient Gradient Inversion with Fine-Grained Reconstruction cites Inverting Gradients - How easy is it to break privacy in federated learning?
  • Cracking Federated Privacy: Initialization-Resilient Gradient Inversion with Fine-Grained Reconstruction cites Deep Learning with Differential Privacy
  • SoK: Colluding Adversaries in Machine Learning Pipelines cites Inverting Gradients - How easy is it to break privacy in federated learning?
  • On the Detectability of Active Gradient Inversion Attacks in Federated Learning cites Inverting Gradients - How easy is it to break privacy in federated learning?
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers4

  1. On the Detectability of Active Gradient Inversion Attacks in Federated LearningVincenzo Carletti, Pasquale Foggia, Carlo Mazzocca, Giuseppe Parrella et al.S&P 2026 · 1 citation
  2. When Topology Betrays Privacy: Lattice-Based Reconstruction Attacks on Secure Aggregation in Decentralized Federated LearningWenrui Yu, Changlong Ji, Johannes Bjerva, Qiongxiu LiCCS 2026
  3. Cracking Federated Privacy: Initialization-Resilient Gradient Inversion with Fine-Grained ReconstructionKaiming Zhu, Jinsheng Yang, Siyang Guo, Huaqian Qin et al.USENIX Security 2026
  4. SoK: Colluding Adversaries in Machine Learning PipelinesVasisht Duddu, Lipeng He, Asim Waheed, N. AsokanUSENIX Security 2026

Builds on44

  1. Deep Learning with Differential PrivacyMartín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan et al.CCS 2016 · 7,620 citations
  2. Practical Secure Aggregation for Privacy-Preserving Machine LearningKallista A. Bonawitz, Vladimir Ivanov, Ben Kreuter, Antonio Marcedone et al.CCS 2017 · 3,936 citations
  3. SCAFFOLD: Stochastic Controlled Averaging for Federated LearningSai Praneeth Karimireddy, Satyen Kale, Mehryar Mohri, Sashank J. Reddi et al.ICML 2020 · 3,875 citations
  4. Inverting Gradients - How easy is it to break privacy in federated learning?Jonas Geiping, Hartmut Bauermeister, Hannah Dröge, Michael MoellerNeurIPS 2020 · 1,822 citations
  5. Deep Models Under the GAN: Information Leakage from Collaborative Deep LearningBriland Hitaj, Giuseppe Ateniese, Fernando Pérez-CruzCCS 2017 · 1,581 citations

Related papers

  1. SoK: On Gradient Leakage in Federated LearningJiacheng Du, Jiahui Hu, Zhibo Wang, Peng Sun et al.USENIX Security 2025
  2. Evaluating Gradient Inversion Attacks and Defenses in Federated LearningYangsibo Huang, Samyak Gupta, Zhao Song, Kai Li et al.NeurIPS 2021 · 419 citations
  3. Leak and Learn: An Attacker's Cookbook to Train Using Leaked Data from Federated LearningJoshua C. Zhao, Ahaan Dabholkar, Atul Sharma, Saurabh BagchiCVPR 2024 · 4 citations
  4. Soteria: Provable Defense Against Privacy Leakage in Federated Learning From Representation PerspectiveJingwei Sun, Ang Li, Binghui Wang, Huanrui Yang et al.CVPR 2021
  5. SVDefense: Effective Defense against Gradient Inversion Attacks via Singular Value DecompositionChenxiang Luo, David K. Y. Yau, Qun SongNDSS 2026 · 3 citations
Dusk over the sea between two cliffs drawn in fine vertical lines