mlsanityData to 14 Oct 2026

DP-MIND: Differentially private memory-guided Long context Inference

arXiv 2610.13132 ↗First posted 9 Oct 2026cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

LLM-as-a-Service has become a dominant paradigm for accessing large language models without requiring local deployment. However, the process raises severe privacy concerns when user queries may include sensitive or confidential contexts. Differential privacy (DP) provides a principled approach to protecting such inputs, but existing DP inference methods are primarily designed for short- or moderate-context settings and struggle with million-token-scale inputs due to limited context capacity and severe utility degradation under DP perturbation. In this paper, we propose DP-MIND, a memory-guided framework that scales differentially private inference to million-token contexts. DP-MIND enables private long-context inference by locally perturbing the input and processing it chunk by chunk with a compact memory that preserves task-relevant information across the sequence. To make this memory robust to DP noise, we optimize the memory construction policy directly on perturbed inputs using reinforcement learning, guided by both final answer correctness and memory consistency. We further recover perturbation-corrupted answer spans through a local denoising module without exposing the original context to the external model. We provide theoretical privacy analysis of DP-MIND and further conduct extensive empirical analysis across multiple long-context benchmarks with context lengths up to the million-token scale, demonstrating that DP-MIND consistently outperforms existing DP and long-context inference baselines in both scalability and utility.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 27 of its 34 references that are in the map. First posted on arXiv 9 Oct 2026.

Builds on

27 of 34 references.

References · 27

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  4. 2024Qwen2.5 Technical Reportcited by 4,942
  5. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  6. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  7. 2020Longformer: The Long-Document Transformercited by 3,413
  8. 2023A Survey of Large Language Modelscited by 2,576
  9. 2020Big Bird: Transformers for Longer Sequencescited by 1,684
  10. 2024RULER: What's the Real Context Size of Your Long-Context Language Models?cited by 1,346
  11. 2023Extending Context Window of Large Language Models via Positional Interpolationcited by 740
  12. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 601
15 more
  1. 2024LongRoPE: Extending LLM Context Window Beyond 2 Million Tokenscited by 313
  2. 2025MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agentcited by 246
  3. 2025Qwen2.5-1M Technical Reportcited by 164
  4. 2019Privacy- and Utility-Preserving Textual Analysis via Calibrated Multivariate Perturbationscited by 119
  5. 2021Differential Privacy for Text Analytics via Natural Text Sanitizationcited by 77
  6. 2023DP-Forward: Fine-tuning and Inference on Language Models with Differential Privacy in Forward Passcited by 67
  7. 2022A Customized Text Sanitization Mechanism with Differential Privacycited by 52
  8. 2025Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Surveycited by 51
  9. 2020A Differentially Private Text Perturbation Method Using a Regularized Mahalanobis Metriccited by 42
  10. 2023Split-and-Denoise: Protect large language model inference with local differential privacycited by 32
  11. 2025A Survey on Privacy Risks and Protection in Large Language Modelscited by 21
  12. 2023InferDPT: Privacy-Preserving Inference for Closed-box Large Language Modelcited by 8
  13. 2026Dynamic Long Context Reasoning over Compressed Memory via End-to-End Reinforcement Learningcited by 8
  14. 2025DP-Fusion: Token-Level Differentially Private Inference for Large Language Modelscited by 6
  15. 2026SharedRequest: Privacy-Preserving Model-Agnostic Inference for Large Language Modelscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026DP-OPD: Differentially Private On-Policy Distillation for Language Models
  2. 2024PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restoration
  3. 2025Public Data Assisted Differentially Private In-Context Learning
  4. 2025Comet: Accelerating Private Inference for Large Language Model by Predicting Activation Sparsity
  5. 2026SpliTEE: Fast and Private LLM Inference by Coupling GPU-Assisted Trusted Execution Environments with Differential Privacy
  6. 2025Towards Confidential and Efficient LLM Inference with Dual Privacy Protection
  7. 2024Efficient and Private: Memorisation under differentially private parameter-efficient fine-tuning in language models
  8. 2025Dual-Priv Pruning : Efficient Differential Private Fine-Tuning in Multimodal Large Language Models