Data to 7 Oct 2026

iDLG: Improved Deep Leakage from Gradients

arXiv 2001.02610 ↗First posted 8 Jan 2020cs.LG stat.MLcited by 367

Not on the map. Why?

Abstract (arXiv)

It is widely believed that sharing gradients will not leak private training data in distributed learning systems such as Collaborative Learning and Federated Learning, etc. Recently, Zhu et al. presented an approach which shows the possibility to obtain private training data from the publicly shared gradients. In their Deep Leakage from Gradient (DLG) method, they synthesize the dummy data and corresponding labels with the supervision of shared gradients. However, DLG has difficulty in convergence and discovering the ground-truth labels consistently. In this paper, we find that sharing gradients definitely leaks the ground-truth labels. We propose a simple but reliable approach to extract accurate data from the gradients. Particularly, our approach can certainly extract the ground-truth labels as opposed to DLG, hence we name it Improved DLG (iDLG). Our approach is valid for any differentiable model trained with cross-entropy loss over one-hot labels. We mathematically illustrate how our method can extract ground-truth labels from the gradients and empirically demonstrate the advantages over DLG.

Key facts. Cited by 367 papers in the map (data through 7 Oct 2026). First posted on arXiv 8 Jan 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 44; 2022: 65; 2023: 84; 2024: 79; 2025: 58; 2026: 37.

367 papers.

Sort:

By area:

2021Federated Learning on Non-IID Data: A Surveycited by 345
2021See through Gradients: Image Batch Recovery via GradInversioncited by 320
2021Evaluating Gradient Inversion Attacks and Defenses in Federated Learningcited by 196
2023Text Embeddings Reveal (Almost) As Much As Textcited by 189
2021When the Curious Abandon Honesty: Federated Learning Is Not Privatecited by 126
2022Federated Class-Incremental Learningcited by 126
2021Robbing the Fed: Directly Obtaining Private Data in Federated Learning with Modified Modelscited by 106
2021Gradient Inversion with Generative Image Priorcited by 101
2021Trustworthy AI: A Computational Perspectivecited by 93
2021Label Leakage and Protection in Two-party Split Learningcited by 87
2022Survey on Federated Learning Threats: concepts, taxonomy on attacks and defences, experimental study and challengescited by 81
2023Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentencecited by 81

38 more

2021CAFE: Catastrophic Data Leakage in Vertical Federated Learningcited by 76
2022Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakagecited by 72
2022Recovering Private Text in Federated Learning of Language Modelscited by 67
2021Understanding Clipping for Federated Learning: Convergence and Client-Level Differential Privacycited by 66
2021From Distributed Machine Learning to Federated Learning: A Surveycited by 65
2021Privacy-Preserving Machine Learning: Methods, Challenges and Directionscited by 58
2022THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryptioncited by 55
2022FederatedScope: A Flexible Federated Learning Platform for Heterogeneitycited by 53
2023Federated Continual Learning via Knowledge Fusion: A Surveycited by 53
2021UnSplit: Data-Oblivious Model Inversion, Model Stealing, and Label Inference Attacks Against Split Learningcited by 52
2022LAMP: Extracting Text from Gradients with Language Model Priorscited by 52
2023Privacy in Large Language Models: Attacks, Defenses and Future Directionscited by 48
2023GIFD: A Generative Gradient Inversion Method with Feature Domain Optimizationcited by 45
2021FLOP: Federated Learning on Medical Datasets using Partial Networkscited by 42
2023Bounding Training Data Reconstruction in DP-SGDcited by 42
2022Closing the Generalization Gap of Cross-silo Federated Medical Image Segmentationcited by 41
2021Towards General Deep Leakage in Federated Learningcited by 40
2023Recent Advances on Federated Learning: A Systematic Surveycited by 40
2021FedCG: Leverage Conditional GAN for Protecting Privacy and Maintaining Competitive Performance in Federated Learningcited by 39
2022No Free Lunch Theorem for Security and Utility in Federated Learningcited by 39
2022A Survey on Gradient Inversion: Attacks, Defenses and Future Directionscited by 39
2022Gradient Obfuscation Gives a False Sense of Security in Federated Learningcited by 37
2021GRNN: Generative Regression Neural Network -- A Data Leakage Attack for Federated Learningcited by 35
2023LOKI: Large-scale Data Reconstruction Attack against Federated Learning through Model Manipulationcited by 35
2021Bayesian Framework for Gradient Leakagecited by 34
2022Joint Privacy Enhancement and Quantization in Federated Learningcited by 34
2022Trading Off Privacy, Utility and Efficiency in Federated Learningcited by 33
2024Advances and Open Challenges in Federated Foundation Modelscited by 33
2022Learning to Invert: Simple Adaptive Attacks for Gradient Inversion in Federated Learningcited by 30
2021Dynamic Differential-Privacy Preserving SGDcited by 29
2022Cocktail Party Attack: Breaking Aggregation-Based Privacy in Federated Learning using Independent Component Analysiscited by 29
2021PRECODE - A Generic Model Extension to Prevent Deep Gradient Leakagecited by 28
2022Reconstructing Training Data from Model Gradient, Provablycited by 27
2023A Survey on Decentralized Federated Learningcited by 27
2022Data Leakage in Federated Averagingcited by 26
2022CrowdGuard: Federated Backdoor Detection in Federated Learningcited by 26
2023Privacy and Fairness in Federated Learning: on the Perspective of Trade-offcited by 26
2022Sketching for First Order Method: Efficient Algorithm for Low-Bandwidth Channel and Vulnerabilitycited by 25

50 most cited of 367 All 367 (JSON)