mlsanityData to 14 Oct 2026

Federated Learning of Gboard Language Models with Differential Privacy

arXiv 2305.18465 ↗First posted 29 May 2023cs.LG cs.CRcited by 66

Not on the map. Why?

Abstract (arXiv)

We train language models (LMs) with federated learning (FL) and differential privacy (DP) in the Google Keyboard (Gboard). We apply the DP-Follow-the-Regularized-Leader (DP-FTRL)~citepkairouz21b algorithm to achieve meaningfully formal DP guarantees without requiring uniform sampling of client devices. To provide favorable privacy-utility trade-offs, we introduce a new client participation criterion and discuss the implication of its configuration in large scale systems. We show how quantile-based clip estimation~citepandrew2019differentially can be combined with DP-FTRL to adaptively choose the clip norm during training or reduce the hyperparameter tuning in preparation for training. With the help of pretraining on public data, we train and deploy more than twenty Gboard LMs that achieve high utility and ρ-zCDP privacy guarantees with ρ ∈ (0.2, 2), with two models additionally trained with secure aggregation~citepbonawitz2017practical. We are happy to announce that all the next word prediction neural network LMs in Gboard now have DP guarantees, and all future launches of Gboard neural network LMs will require DP guarantees. We summarize our experience and provide concrete suggestions on DP training for practitioners.

Key facts. Cited by 66 papers in the map (data through 14 Oct 2026). Builds on 28 of its 46 references that are in the map. First posted on arXiv 29 May 2023.

Builds on

28 of 46 references.

References · 28

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  3. 2019Advances and Open Problems in Federated Learningcited by 3,046
  4. 2016Deep Learning with Differential Privacycited by 2,728
  5. 2020mT5: A massively multilingual pre-trained text-to-text transformercited by 1,818
  6. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 864
  7. 2019Towards Federated Learning at Scale: System Designcited by 832
  8. 2018Federated Learning for Mobile Keyboard Predictioncited by 572
  9. 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 459
  10. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 412
  11. 2021Differentially Private Fine-tuning of Language Modelscited by 400
  12. 2016Concentrated Differential Privacy: Simplifications, Extensions, and Lower Boundscited by 346
16 more
  1. 2021A Field Guide to Federated Optimizationcited by 315
  2. 2022Unlocking High-Accuracy Differentially Private Image Classification through Scalecited by 224
  3. 2019Differentially Private Learning with Adaptive Clippingcited by 220
  4. 2018Applied Federated Learning: Improving Google Keyboard Query Suggestionscited by 204
  5. 2021Practical and Private (Deep) Learning without Sampling or Shufflingcited by 170
  6. 2023How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacycited by 153
  7. 2021The Distributed Discrete Gaussian Mechanism for Federated Learning with Secure Aggregationcited by 128
  8. 2020Training Production Language Models without Memorizing User Datacited by 68
  9. 2021The Skellam Mechanism for Differentially Private Federated Learningcited by 68
  10. 2022Multi-Epoch Matrix Factorization Mechanisms for Private Machine Learningcited by 48
  11. 2020Privacy Amplification via Random Check-Inscited by 39
  12. 2022The Fundamental Price of Secure Aggregation in Differentially Private Federated Learningcited by 39
  13. 2017Compressing Word Embeddings via Deep Compositional Code Learningcited by 35
  14. 2023On the Convergence of Federated Averaging with Cyclic Client Participationcited by 31
  15. 2023Can Public Large Language Models Help Private Cross-device Federated Learning?cited by 30
  16. 2022Learning to Generate Image Embeddings with User-level Differential Privacycited by 23

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 16; 2024: 26; 2025: 17; 2026: 7.

66 papers.

Sort:

By area:

  1. 2025Multi-Agent Collaboration Mechanisms: A Survey of LLMscited by 431
  2. 2023(Amplified) Banded Matrix Factorization: A unified approach to private trainingcited by 49
  3. 2023User Inference Attacks on Large Language Modelscited by 36
  4. 2025Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agentscited by 36
  5. 2024Teach LLMs to Phish: Stealing Private Information from Language Modelscited by 32
  6. 2024A Survey on Efficient Federated Learning Methods for Foundation Model Trainingcited by 29
  7. 2024PrE-Text: Training Language Models on Private Federated Data in the Age of LLMscited by 29
  8. 2024Fine-Tuning Large Language Models with User-Level Differential Privacycited by 29
  9. 2023How to Protect Copyright Data in Optimization of Large Language Models?cited by 26
  10. 2023Grounding Foundation Models through Federated Transfer Learning: A General Frameworkcited by 24
  11. 2023Correlated Noise Provably Beats Independent Noise for Differentially Private Learningcited by 23
  12. 2023Enabling On-Device Large Language Model Personalization with Self-Supervised Data Selection and Synthesiscited by 20
38 more
  1. 2024Differentially Private Knowledge Distillation via Synthetic Text Generationcited by 18
  2. 2024Prompt Public Large Language Models to Synthesize Data for Private On-device Applicationscited by 17
  3. 2024Efficient and Near-Optimal Noise Generation for Streaming Differential Privacycited by 17
  4. 2024Empirical Guidelines for Deploying LLMs onto Resource-constrained Edge Devicescited by 17
  5. 2025Synthesizing Privacy-Preserving Text Data via Finetuning without Finetuning Billion-Scale LLMscited by 16
  6. 2024Federated Learning in Practice: Reflections and Projectionscited by 15
  7. 2023Private Distribution Learning with Public Data: The View from Sample Compressioncited by 14
  8. 2024Synergizing Foundation Models and Federated Learning: A Surveycited by 14
  9. 2025POPri: Private Federated Learning using Preference-Optimized Synthetic Datacited by 14
  10. 2024The Federation Strikes Back: A Survey of Federated Learning Privacy Attacks, Defenses, Applications, and Policy Landscapecited by 13
  11. 2023Unmasking Transformers: A Theoretical Approach to Data Recovery via Attention Weightscited by 12
  12. 2024Private Gradient Descent for Linear Regression: Tighter Error Bounds and Instance-Specific Uncertainty Estimationcited by 12
  13. 2023Information Flow Control in Machine Learning through Modular Model Architecturecited by 11
  14. 2023FedTherapist: Mental Health Monitoring with User-Generated Linguistic Expressions on Smartphones via Federated Learningcited by 10
  15. 2024DP-DyLoRA: Fine-Tuning Transformer-Based Models On-Device under Differentially Private Federated Learning using Dynamic Low-Rank Adaptationcited by 9
  16. 2024A Hassle-free Algorithm for Private Learning in Practice: Don't Use Tree Aggregation, Use BLTscited by 9
  17. 2024pfl-research: simulation framework for accelerating research in Private Federated Learningcited by 8
  18. 2024Thinking Racial Bias in Fair Forgery Detection: Models, Datasets and Evaluationscited by 8
  19. 2024A Statistical Viewpoint on Differential Privacy: Hypothesis Testing, Representation and Blackwell's Theoremcited by 8
  20. 2023When Federated Learning meets Watermarking: A Comprehensive Overview of Techniques for Intellectual Property Protectioncited by 7
  21. 2024Byzantine-Robust Aggregation for Securing Decentralized Federated Learningcited by 7
  22. 2025Mitigating Unintended Memorization with LoRA in Federated Learning for LLMscited by 7
  23. 2023Samplable Anonymous Aggregation for Private Federated Data Analysiscited by 6
  24. 2023A Theoretical Insight into Attack and Defense of Gradient Leakage in Transformercited by 6
  25. 2024On the Convergence of DP-SGD with Adaptive Clippingcited by 6
  26. 2023Private Federated Learning with Autotuned Compressioncited by 5
  27. 2024Confidential Federated Computationscited by 4
  28. 2024Improved Communication-Privacy Trade-offs in L2 Mean Estimation under Streaming Differential Privacycited by 4
  29. 2024Momentum Approximation in Asynchronous Private Federated Learningcited by 3
  30. 2024Fingerprinting Codes Meet Geometry: Improved Lower Bounds for Private Query Release and Adaptive Data Analysiscited by 3
  31. 2025Empirical Privacy Variancecited by 3
  32. 2025ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted Controlcited by 3
  33. 2024Efficient Language Model Architectures for Differentially Private Federated Learningcited by 2
  34. 2025Synthesizing and Adapting Error Correction Data for Mobile Large Language Model Applicationscited by 2
  35. 2025Constant Stepsize Local GD for Logistic Regression: Acceleration by Instabilitycited by 2
  36. 2026LLM Agents Enable User-Governed Personalization Beyond Platform Boundariescited by 2
  37. 2023Robust and Actively Secure Serverless Collaborative Learningcited by 1
  38. 2023Venn: Resource Management for Collaborative Learning Jobscited by 1

50 most cited of 66 All 66 (JSON)