mlsanityData to 14 Oct 2026

LLMs Can Understand Encrypted Prompt: Towards Privacy-Computing Friendly Transformers

arXiv 2305.18396 ↗First posted 28 May 2023cs.LG cs.CL cs.CRcited by 22

Not on the map. Why?

Abstract (arXiv)

The community explored to build private inference frameworks for transformer-based large language models (LLMs) in a server-client setting, where the server holds the model parameters and the client inputs its private data (or prompt) for inference. However, these frameworks impose significant overhead when the private inputs are forward propagated through the original LLMs. In this paper, we show that substituting the computation- and communication-heavy operators in the transformer architecture with privacy-computing friendly approximations can greatly reduce the private inference costs while incurring very minor impact on model performance. Compared to state-of-the-art Iron (NeurIPS 2022), our privacy-computing friendly model inference pipeline achieves a 5× acceleration in computation and an 80% reduction in communication overhead, while retaining nearly identical accuracy.

Key facts. Cited by 22 papers in the map (data through 14 Oct 2026). Builds on 20 of its 46 references that are in the map. First posted on arXiv 28 May 2023.

Builds on

20 of 46 references.

References · 20

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  4. 2020Language Models are Few-Shot Learnerscited by 32,419
  5. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  6. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  7. 2016Layer Normalizationcited by 5,669
  8. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,899
  9. 2016Gaussian Error Linear Units (GELUs)cited by 3,502
  10. 2019Are Sixteen Heads Really Better than One?cited by 816
  11. 2019Patient Knowledge Distillation for BERT Model Compressioncited by 508
  12. 2018Gazelle: A Low Latency Framework for Secure Neural Network Inferencecited by 193
8 more
  1. 2021Block Pruning For Faster Transformerscited by 177
  2. 2021CrypTen: Secure Multi-Party Computation Meets Machine Learningcited by 131
  3. 2019Well-Read Students Learn Better: On the Importance of Pre-training Compact Modelscited by 120
  4. 2020CrypTFlow2: Practical 2-Party Secure Inferencecited by 82
  5. 2022MPCFormer: fast, performant and private Transformer inference with MPCcited by 69
  6. 2021Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristicscited by 67
  7. 2022THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryptioncited by 57
  8. 2021CryptGPU: Fast Privacy-Preserving Machine Learning on the GPUcited by 46

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 6; 2024: 8; 2025: 4; 2026: 4.

22 papers.

Sort:

By area:

  1. 2023MasterKey: Automated Jailbreak Across Multiple Large Language Model Chatbotscited by 183
  2. 2023Split-and-Denoise: Protect large language model inference with local differential privacycited by 32
  3. 2024PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restorationcited by 28
  4. 2023Grounding Foundation Models through Federated Transfer Learning: A General Frameworkcited by 24
  5. 2023Converting Transformers to Polynomial Form for Secure Inference Over Homomorphic Encryptioncited by 19
  6. 2024No Free Lunch Theorem for Privacy-Preserving LLM Inferencecited by 13
  7. 2023East: Efficient and Accurate Secure Transformer Framework for Inferencecited by 9
  8. 2024EmojiPrompt: Generative Prompt Obfuscation for Privacy-Preserving Communication with Cloud-based LLMscited by 9
  9. 2024Power-Softmax: Towards Secure LLM Inference over Encrypted Datacited by 7
  10. 2025LLM Access Shield: Domain-Specific LLM Framework for Privacy Policy Compliancecited by 6
  11. 2024ObfuscaTune: Obfuscated Offsite Fine-tuning and Inference of Proprietary LLMs on Private Datasetscited by 4
  12. 2025Comet: Accelerating Private Inference for Large Language Model by Predicting Activation Sparsitycited by 4
10 more
  1. 2023ConfusionPrompt: Practical Private Inference for Online Large Language Modelscited by 3
  2. 2024Data-Centric AI in the Age of Large Language Modelscited by 2
  3. 2025Private Transformer Inference in MLaaS: A Surveycited by 2
  4. 2026Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Promptcited by 2
  5. 2024Pencil: Private and Extensible Collaborative Learning without the Non-Colluding Assumptioncited by 1
  6. 2024Confidential Prompting: Privacy-preserving LLM Inference on Cloudcited by 1
  7. 2025A Survey: Towards Privacy and Security in Mobile Large Language Modelscited by 1
  8. 2026Your Inference Request Will Become a Black Box: Confidential Inference for Cloud-based Large Language Modelscited by 1
  9. 2026Continual Pretraining on Encrypted Synthetic Data for Privacy-Preserving LLMscited by 0
  10. 2026A Graph-Based Analysis of Comparability, Scale, and Attribution in Private Transformer Inferencecited by 0