mlsanityData to 7 Oct 2026

Synthetic Query Generation for Privacy-Preserving Deep Retrieval Systems using Differentially Private Language Models

arXiv 2305.05973 ↗First posted 10 May 2023cs.CL cs.CR cs.IRcited by 10

Not on the map. Why?

Abstract (arXiv)

We address the challenge of ensuring differential privacy (DP) guarantees in training deep retrieval systems. Training these systems often involves the use of contrastive-style losses, which are typically non-per-example decomposable, making them difficult to directly DP-train with since common techniques require per-example gradients. To address this issue, we propose an approach that prioritizes ensuring query privacy prior to training a deep retrieval system. Our method employs DP language models (LMs) to generate private synthetic queries representative of the original data. These synthetic queries can be used in downstream retrieval system training without compromising privacy. Our approach demonstrates a significant enhancement in retrieval quality compared to direct DP-training, all while maintaining query-level privacy guarantees. This work highlights the potential of harnessing LMs to overcome limitations in standard DP-training methods.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 35 of its 42 references that are in the map. First posted on arXiv 10 May 2023.

Builds on

35 of 42 references.

References · 35

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  4. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  5. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  6. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  7. 2016Deep Learning with Differential Privacycited by 2,682
  8. 2020Extracting Training Data from Large Language Modelscited by 2,144
  9. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
  10. 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,591
  11. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,225
  12. 2018Adafactor: Adaptive Learning Rates with Sublinear Memory Costcited by 896
23 more
  1. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 847
  2. 2021Membership Inference Attacks From First Principlescited by 839
  3. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754
  4. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 466
  5. 2021Large Dual Encoders Are Generalizable Retrieverscited by 454
  6. 2020RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answeringcited by 410
  7. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 405
  8. 2022Promptagator: Few-shot Dense Retrieval From 8 Examplescited by 233
  9. 2022Unlocking High-Accuracy Differentially Private Image Classification through Scalecited by 218
  10. 2020Embedding-based Retrieval in Facebook Searchcited by 155
  11. 2023How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacycited by 150
  12. 2019Learning Dense Representations for Entity Retrievalcited by 122
  13. 2022Measuring Forgetting of Memorized Training Examplescited by 118
  14. 2022Synthetic Text Generation with Differential Privacy: A Simple and Practical Recipecited by 108
  15. 2021Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?cited by 91
  16. 2019Analyzing Information Leakage of Updates to Natural Language Modelscited by 89
  17. 2020Few-Shot Text Generation with Pattern-Exploiting Trainingcited by 87
  18. 2020Training Production Language Models without Memorizing User Datacited by 68
  19. 2022Differentially Private Language Models for Secure Data Sharingcited by 68
  20. 2020Enabling Fast Differentially Private SGD via Just-in-Time Compilation and Vectorizationcited by 58
  21. 2022Defending against Reconstruction Attacks with R'enyi Differential Privacycited by 30
  22. 2022Privacy-Preserving Domain Adaptation of Semantic Parserscited by 12
  23. 2021Towards Sharper Utility Bounds for Differentially Private Pairwise Learningcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 4; 2026: 4.

10 papers.

  1. 2025Synthesizing Privacy-Preserving Text Data via Finetuning without Finetuning Billion-Scale LLMscited by 15
  2. 2024LLM-based Privacy Data Augmentation Guided by Knowledge Distillation with a Distribution Tutor for Medical Text Classificationcited by 8
  3. 2023Modeling Recommender Ecosystems: Research Challenges at the Intersection of Mechanism Design, Reinforcement Learning and Generative Modelscited by 7
  4. 2025Clustering and Median Aggregation Improve Differentially Private Inferencecited by 5
  5. 2025ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted Controlcited by 2
  6. 2026EPSVec: Efficient and Private Synthetic Data Generation via Dataset Vectorscited by 2
  7. 2025Catalog-Native LLM: Speaking Item-ID Dialect with Less Entanglement for Recommendationcited by 1
  8. 2026Differentially Private Preference Data Synthesis for Large Language Model Alignmentcited by 1
  9. 2026MAPLE: Metadata Augmented Private Language Evolutioncited by 0
  10. 2026PrivBench: A Holistic and Modular Benchmarking Platform for Evaluating Text-to-Text Privatizationcited by 0