Browse by topic¶
Mentions in full text¶
Papers ordered by how often the term occurs (count in brackets). Only papers with a local PDF are counted.
Homomorphic encryption (241 papers)¶
Fully Homomorphic Encryption with Chosen-Ciphertext Security from LWE (298), Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities (252), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (244), Fully Homomorphic Encryption for Machine Learning (222), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (212), Threshold Cryptosystems From Threshold Fully Homomorphic Encryption (205), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (202), A comprehensive survey on secure outsourced computation and its applications (188), The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption (178), A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption (175), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (174), FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System (164), Secure numerical simulations using fully homomorphic encryption (160), Pasta: A Case for Hybrid Homomorphic Encryption (159), HE-MAN – Homomorphically Encrypted MAchine learning with oNnx models (159), Efficiency Optimization Techniques in Privacy-Preserving Federated Learning With Homomorphic Encryption: A Brief Survey (156), EncFormer: Secure and Efficient Transformer Inference over Encrypted Data (154), On-the-Fly Multiparty Computation on the Cloud via Multikey Fully Homomorphic Encryption (153), Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption (151), Private machine learning classification based on fully homomorphic encryption (148), COMPARISON OF FULLY HOMOMORPHIC ENCRYPTION AND GARBLED CIRCUITS APPROACHES IN PRIVACY-PRESERVING MACHINE LEARNING Kalyan Cheerla Thesis Prepared for the Degree of MASTER OF SCIENCE UNIVERSITY OF NORTH TEXAS July 2025 (146), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (141), Revisiting Module Lattice-based Homomorphic Encryption and Application to Secure-MPC (140), F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption (134), Security guidelines for implementing homomorphic encryption (131), Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering (131), ARK: Fully Homomorphic Encryption Accelerator with Runtime Data Generation and Inter-Operation Key Reuse (131), Activate Me!: Designing Efficient Activation Functions for Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (130), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (129), BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption (128), Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning (125), SoK: Private transformer-based model inference (123), A scalable multi-GPU framework for encrypted large-model inference (123), Encryption-friendly LLM architecture (122), (Leveled) Fully Homomorphic Encryption without Bootstrapping (118), ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour (117), Over 100x Faster Bootstrapping in Fully Homomorphic Encryption through Memory-centric Optimization with GPUs (116), OpenFHE: Open-Source Fully Homomorphic Encryption Library (113), TensorFHE: Achieving Practical Computation on Encrypted Data Using GPGPU (112), A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (110), Efficient Fully Homomorphic Encryption from (Standard) LWE (104), Homomorphic Encryption from Learning with Errors: Conceptually-Simpler, Asymptotically-Faster, Attribute-Based (103), Stream ciphers: A Practical Solution for Efficient Homomorphic-Ciphertext Compression (101), BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption (101), EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation (98), Private and reliable neural network inference (95), Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption (93), Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption (92), MOAI: Module-optimizing architecture for non-interactive secure transformer inference (91), Efficient Homomorphic Evaluation on Large Intervals (90), HHEML: Hybrid Homomorphic Encryption for Privacy-Preserving Machine Learning on Edge (89), GG-GSW: Chosen-Ciphertext Secure Leveled FHE From Gadget Trapdoors (86), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (85), Federated Learning: An approach with Hybrid Homomorphic Encryption (85), Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption (84), Privft: Private and fast text classification with homomorphic encryption (83), Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach (83), HeLayers: A Tile Tensors Framework for Large Neural Networks on Encrypted Data (83), Securing Approximate Homomorphic Encryption Using Differential Privacy (83), Revisiting Shamir Secret Sharing for Threshold Fully Homomorphic Encryption (82) …
Open source (203 papers)¶
SoK: Privacy-preserving computation techniques for deep learning (41), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (35), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (28), SoK: Private transformer-based model inference (25), OpenFHE: Open-Source Fully Homomorphic Encryption Library (21), SoK: Private LLM Inference using Approximate Homomorphic Encryption (21), FIDESlib: A Fully-Fledged Open-Source FHE Library for Efficient CKKS on GPUs (19), Security guidelines for implementing homomorphic encryption (18), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (15), MPC-Minimized Secure LLM Inference (15), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (13), HEonGPU: a GPU-based Fully Homomorphic Encryption Library 1.0 (13), Large language model inference acceleration: A comprehensive hardware perspective (11), The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption (11), MOAI: Module-optimizing architecture for non-interactive secure transformer inference (11), One server for the price of two: Simple and fast single-server private information retrieval (10), EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption (10), On the precision loss in approximate homomorphic encryption (9), Secure numerical simulations using fully homomorphic encryption (9), MP-SPDZ: A Versatile Framework for Multi-Party Computation (9), SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine Learning (9), Multiparty Homomorphic Encryption from Ring-Learning-with-Errors (9), ARION: Attention-optimized transformer inference on encrypted data (9), ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs (8), Transciphering Framework for Approximate Homomorphic Encryption (8), HE-MAN – Homomorphically Encrypted MAchine learning with oNnx models (8), Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption (8), DaCapo: Automatic Bootstrapping Management for Efficient Fully Homomorphic Encryption (8), Orion: A Fully Homomorphic Encryption Framework for Deep Learning (8), Nimbus: Secure and Efficient Two-Party Inference for Transformers (8), BumbleBee: Secure Two-party Inference Framework for Large Transformers (8), Efficient Bootstrapping for Approximate Homomorphic Encryption with Non-Sparse Keys (8), Practical secure inference algorithm for a fine-tuned large language model based on fully homomorphic encryption (8), Theory, analysis, and best practices for sigmoid selfattention (7), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (7), Rubato : Noisy Ciphers for Approximate Homomorphic Encryption (Full Version) (7), Evaluation of Privacy-aware Support Vector Machine (SVM) Learning using Homomorphic Encryption (7), EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation (7), Phantom: A CUDA-Accelerated Word-Wise Homomorphic Encryption Library (7), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (7), CipherGPT: Secure Two-Party GPT Inference (7), Cheetah: Lean and Fast Secure Two-Party Deep Neural Network Inference (7), Securing Approximate Homomorphic Encryption Using Differential Privacy (7), Llama: Open and efficient foundation language models (6), Fully Homomorphic Encryption for Cyclotomic Prime Moduli (6), COMPARISON OF FULLY HOMOMORPHIC ENCRYPTION AND GARBLED CIRCUITS APPROACHES IN PRIVACY-PRESERVING MACHINE LEARNING Kalyan Cheerla Thesis Prepared for the Degree of MASTER OF SCIENCE UNIVERSITY OF NORTH TEXAS July 2025 (6), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (6), Federated Learning: An approach with Hybrid Homomorphic Encryption (6), Private pathological assessment via machine learning and homomorphic encryption (6), CrypTFlow2: Practical 2-Party Secure Inference (6), On the Security of Homomorphic Encryption on Approximate Numbers (6), An Open-Source End-to-End FHE Implementation for Privacy-Preserving Llama 3 8B Inference (6), ENSI: Efficient non-interactive secure inference for large language models (6), Homomorphic encryption standard (5), Private and reliable neural network inference (5), Towards Case-Optimized Hybrid Homomorphic Encryption (5), Ajax: Fast Threshold Fully Homomorphic Encryption without Noise Flooding (5), Revisiting Shamir Secret Sharing for Threshold Fully Homomorphic Encryption (5), FHE for SIMD Arithmetic Logic Units with Amortized O(1) Bootstrapping per Ciphertext (5), Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning (5) …
Benchmark (171 papers)¶
Pasta: A Case for Hybrid Homomorphic Encryption (83), PrivateLoRA for Efficient Privacy Preserving LLM (34), EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption (32), Llama: Open and efficient foundation language models (31), CrypTFlow2: Practical 2-Party Secure Inference (28), GAZELLE: A Low Latency Framework for Secure Neural Network Inference (24), F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption (20), SoK: Private LLM Inference using Approximate Homomorphic Encryption (18), DaCapo: Automatic Bootstrapping Management for Efficient Fully Homomorphic Encryption (17), Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware (17), CrypTen: Secure Multi-Party Computation Meets Machine Learning (17), BASALISC: Programmable Hardware Accelerator for BGV Fully Homomorphic Encryption (16), MPCFormer: Fast, Performant and Private Transformer Inference with MPC (14), A scalable multi-GPU framework for encrypted large-model inference (14), Theory, analysis, and best practices for sigmoid selfattention (13), MP-SPDZ: A Versatile Framework for Multi-Party Computation (13), ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs (12), GLUE: A multi-task benchmark and analysis platform for natural language understanding (12), FIDESlib: A Fully-Fledged Open-Source FHE Library for Efficient CKKS on GPUs (12), STIP: Efficient and secure non-interactive transformer inference via compact packing (12), One server for the price of two: Simple and fast single-server private information retrieval (11), MPC-Minimized Secure LLM Inference (11), Towards Case-Optimized Hybrid Homomorphic Encryption (10), The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption (10), HeLayers: A Tile Tensors Framework for Large Neural Networks on Encrypted Data (10), CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption (10), Secure transformer inference made non-interactive (10), [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md) (9), HellaSwag: Can a machine really finish your sentence? (9), Improving language understanding by generative pretraining (9), Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping (9), Stay in Your Lane: Fast Arithmetic over Large Finite Rings with CKKS (9), CipherGPT: Secure Two-Party GPT Inference (9), Cheetah: Lean and Fast Secure Two-Party Deep Neural Network Inference (9), PEGASUS: Bridging Polynomial and Non-polynomial Evaluations in Homomorphic Encryption (9), General Functional Bootstrapping using CKKS (9), ARION: Attention-optimized transformer inference on encrypted data (9), BERT: Pre-training of deep bidirectional transformers for language understanding (8), Demystifying bootstrapping in fully homomorphic encryption (8), COMPARISON OF FULLY HOMOMORPHIC ENCRYPTION AND GARBLED CIRCUITS APPROACHES IN PRIVACY-PRESERVING MACHINE LEARNING Kalyan Cheerla Thesis Prepared for the Degree of MASTER OF SCIENCE UNIVERSITY OF NORTH TEXAS July 2025 (8), Non-interactive Fully Encrypted Machine Learning Protocol for Inference (8), Large language models in medicine (7), Phantom: A CUDA-Accelerated Word-Wise Homomorphic Encryption Library (7), SHAFT: Secure, Handy, Accurate, and Fast Transformer Inference (7), Trace-Factored BigSwitch for Matrix-Friendly FHE (7), Revisiting Homomorphic Encryption Schemes for Finite Fields (7), Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity (7), The LAMBADA dataset: Word prediction requiring a broad discourse context (6), Pointer sentinel mixture models (6), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (6), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (6), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (6), TenSEAL: A Library for Encrypted Tensor Operations Using Homomorphic Encryption (6), Orion: A Fully Homomorphic Encryption Framework for Deep Learning (6), Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy (6), BumbleBee: Secure Two-party Inference Framework for Large Transformers (6), SIGMA: Secure GPT Inference with Function Secret Sharing (6), THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption (6), ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation (6), THEMIS: A Co-Designed System for Encrypted Transformer Inference (6) …
Private / secure inference (102 papers)¶
MPC-Minimized Secure LLM Inference (76), SIGMA: Secure GPT Inference with Function Secret Sharing (53), COMPARISON OF FULLY HOMOMORPHIC ENCRYPTION AND GARBLED CIRCUITS APPROACHES IN PRIVACY-PRESERVING MACHINE LEARNING Kalyan Cheerla Thesis Prepared for the Degree of MASTER OF SCIENCE UNIVERSITY OF NORTH TEXAS July 2025 (44), Ditto: Quantization-aware Secure Inference of Transformers upon MPC (39), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (35), SoK: Private transformer-based model inference (35), PUMA: Secure Inference of LLaMA-7B in Five Minutes (32), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (31), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (29), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (27), SHAFT: Secure, Handy, Accurate, and Fast Transformer Inference (26), ENSI: Efficient non-interactive secure inference for large language models (26), Secure human action recognition by encrypted neural network inference (23), CrypTFlow2: Practical 2-Party Secure Inference (23), ARION: Attention-optimized transformer inference on encrypted data (22), Iron: Private Inference on Transformers (20), Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption (18), BumbleBee: Secure Two-party Inference Framework for Large Transformers (18), HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation (17), Secure transformer inference made non-interactive (17), Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache (17), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (14), Evaluation of Privacy-aware Support Vector Machine (SVM) Learning using Homomorphic Encryption (14), Centaur: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference (14), ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation (14), Converting transformers to polynomial form for secure inference over homomorphic encryption (14), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (13), BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers (13), EncFormer: Secure and Efficient Transformer Inference over Encrypted Data (13), SoK: Private LLM Inference using Approximate Homomorphic Encryption (13), Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering (12), TRICYCLE: Private transformer inference with tricyclic encodings (12), Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability (11), Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption (11), CRAfT: Constant-Round Non-Polynomial Approximation and P2P Network for Secure Transformer Inference (11), CipherGPT: Secure Two-Party GPT Inference (10), Swift: Fast Secure Neural Network Inference With Fully Homomorphic Encryption (9), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (9), Terrazzo: Memory-Aware GPU Framework for Private Transformer Inference (9), Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity (9), THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption (8), Delphi: A Cryptographic Inference Service for Neural Networks (8), Practical secure inference algorithm for a fine-tuned large language model based on fully homomorphic encryption (8), Private and reliable neural network inference (7), Activate Me!: Designing Efficient Activation Functions for Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (7), Cheetah: Lean and Fast Secure Two-Party Deep Neural Network Inference (7), THOR: Secure transformer inference with homomorphic encryption (7), Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning (6), HE-MAN – Homomorphically Encrypted MAchine learning with oNnx models (6), Nimbus: Secure and Efficient Two-Party Inference for Transformers (6), MPCFormer: Fast, Performant and Private Transformer Inference with MPC (6), ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour (6), Encryption-friendly LLM architecture (6), Scaling up FHE-based Privacy-Preserving ML: Higher Throughput, Longer Inputs for LLama-3-8B (v1: Scaling up Privacy-Preserving ML: A CKKS Implementation of Llama-2-7B) (6), A scalable multi-GPU framework for encrypted large-model inference (6), Non-interactive Fully Encrypted Machine Learning Protocol for Inference (5), SecFormer: Fast and Accurate Privacy-Preserving Inference for Transformer Models via SMPC (5), Zephyr: GPU-Efficient Homomorphic Encryption for Privacy-Preserving Transformer Inference (5), An Open-Source End-to-End FHE Implementation for Privacy-Preserving Llama 3 8B Inference (5), MOAI: Module-optimizing architecture for non-interactive secure transformer inference (5) …
Privacy-preserving machine learning (99 papers)¶
Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (147), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (102), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (99), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (78), A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption (75), Non-interactive Fully Encrypted Machine Learning Protocol for Inference (67), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (46), Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network (45), Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption (39), HHEML: Hybrid Homomorphic Encryption for Privacy-Preserving Machine Learning on Edge (37), Activate Me!: Designing Efficient Activation Functions for Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (30), SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine Learning (30), COMPARISON OF FULLY HOMOMORPHIC ENCRYPTION AND GARBLED CIRCUITS APPROACHES IN PRIVACY-PRESERVING MACHINE LEARNING Kalyan Cheerla Thesis Prepared for the Degree of MASTER OF SCIENCE UNIVERSITY OF NORTH TEXAS July 2025 (20), SoK: Private transformer-based model inference (19), Recent advances of privacy-preserving machine learning based on (Fully) Homomorphic Encryption (18), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (11), SHAFT: Secure, Handy, Accurate, and Fast Transformer Inference (11), Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption (10), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (9), ABY3: A Mixed Protocol Framework for Machine Learning (9), Practical Considerations of Fully Homomorphic Encryption in Privacy-Preserving Machine Learning (9), Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption (9), CRAfT: Constant-Round Non-Polynomial Approximation and P2P Network for Secure Transformer Inference (9), A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (7), Encryption-friendly LLM architecture (7), Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption (6), Evaluation of Privacy-aware Support Vector Machine (SVM) Learning using Homomorphic Encryption (6), CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption (6), ARION: Attention-optimized transformer inference on encrypted data (6), Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication (5), High-Precision Bootstrapping of RNS-CKKS Homomorphic Encryption Using Optimal Minimax Polynomial Approximation and Inverse Sine Function (5), Converting transformers to polynomial form for secure inference over homomorphic encryption (5), Private and reliable neural network inference (4), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (4), Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data (4), DaCapo: Automatic Bootstrapping Management for Efficient Fully Homomorphic Encryption (4), Ditto: Quantization-aware Secure Inference of Transformers upon MPC (4), BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers (4), PUMA: Secure Inference of LLaMA-7B in Five Minutes (4), Low-Depth Bootstrapping for Matrix-Native FHE (4), Zephyr: GPU-Efficient Homomorphic Encryption for Privacy-Preserving Transformer Inference (4), SoK: Private LLM Inference using Approximate Homomorphic Encryption (4), SoK: Privacy-preserving computation techniques for deep learning (3), Private svm inference on encrypted data (3), Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering (3), Private pathological assessment via machine learning and homomorphic encryption (3), TenSEAL: A Library for Encrypted Tensor Operations Using Homomorphic Encryption (3), Fast and Accurate Homomorphic Softmax Evaluation (3), Secure numerical simulations using fully homomorphic encryption (2), The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption (2), Efficiency Optimization Techniques in Privacy-Preserving Federated Learning With Homomorphic Encryption: A Brief Survey (2), Orion: A Fully Homomorphic Encryption Framework for Deep Learning (2), CrypTen: Secure Multi-Party Computation Meets Machine Learning (2), SIGMA: Secure GPT Inference with Function Secret Sharing (2), PEGASUS: Bridging Polynomial and Non-polynomial Evaluations in Homomorphic Encryption (2), Secure outsourced matrix computation and application to neural networks (1), Cryptanalysis of Symmetric Primitives over Rings and a Key Recovery Attack on Rubato (1), Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping (1), Bootstrapping for BGV and BFV Revisited (1), XBOOT: Free-XOR Gates for CKKS with Applications to Transciphering (1) …
Large language models (67 papers)¶
Large language model inference acceleration: A comprehensive hardware perspective (451), Integrating large language models in financial investments and market analysis: A survey (171), (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice (158), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (149), Large language models in medicine (119), Beyond memorization: Violating privacy via inference with large language models (108), SoK: Private LLM Inference using Approximate Homomorphic Encryption (107), zkLLM: Zero Knowledge Proofs for Large Language Models (91), EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption (84), Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability (68), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (65), Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy (60), Cost-efficient llm training with lifetimeaware tensor offloading via gpudirect storage (54), PrivateLoRA for Efficient Privacy Preserving LLM (50), Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache (44), Practical secure inference algorithm for a fine-tuned large language model based on fully homomorphic encryption (43), ENSI: Efficient non-interactive secure inference for large language models (38), Scaling up FHE-based Privacy-Preserving ML: Higher Throughput, Longer Inputs for LLama-3-8B (v1: Scaling up Privacy-Preserving ML: A CKKS Implementation of Llama-2-7B) (37), ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation (35), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (34), MPC-Minimized Secure LLM Inference (34), TRICYCLE: Private transformer inference with tricyclic encodings (28), Encryption-friendly LLM architecture (27), Bitnet: Scaling 1-bit transformers for large language models (25), SoK: Private transformer-based model inference (25), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (19), GPT-4 System Card (17), ELLMo: Packing- and Depth-Aware Encrypted Transformer Inference (17), A scalable multi-GPU framework for encrypted large-model inference (15), An Open-Source End-to-End FHE Implementation for Privacy-Preserving Llama 3 8B Inference (14), Llama: Open and efficient foundation language models (13), Confidential Computing on NVIDIA Hopper GPUs: A Performance Benchmark Study (11), Plaintext-Ciphertext Matrix Multiplication and FHE Bootstrapping: Fast and Fused (11), ARION: Attention-optimized transformer inference on encrypted data (11), CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption (10), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (9), Error Propagation-Aware Scale Design for Efficient Homomorphic Encryption-Based LLM Inference (9), ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour (7), Theory, analysis, and best practices for sigmoid selfattention (6), CipherGPT: Secure Two-Party GPT Inference (6), Fast and Accurate Homomorphic Softmax Evaluation (6), HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation (6), Zephyr: GPU-Efficient Homomorphic Encryption for Privacy-Preserving Transformer Inference (6), Centaur: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference (5), Nimbus: Secure and Efficient Two-Party Inference for Transformers (5), Terrazzo: Memory-Aware GPU Framework for Private Transformer Inference (5), STIP: Efficient and secure non-interactive transformer inference via compact packing (5), Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity (5), MOAI: Module-optimizing architecture for non-interactive secure transformer inference (4), FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System (3), Ditto: Quantization-aware Secure Inference of Transformers upon MPC (3), SecFormer: Fast and Accurate Privacy-Preserving Inference for Transformer Models via SMPC (3), PUMA: Secure Inference of LLaMA-7B in Five Minutes (3), COMPARISON OF FULLY HOMOMORPHIC ENCRYPTION AND GARBLED CIRCUITS APPROACHES IN PRIVACY-PRESERVING MACHINE LEARNING Kalyan Cheerla Thesis Prepared for the Degree of MASTER OF SCIENCE UNIVERSITY OF NORTH TEXAS July 2025 (2), Fully Homomorphic Encryption for Matrix Arithmetic (2), Converting transformers to polynomial form for secure inference over homomorphic encryption (2), Powerformer: Efficient and high-accuracy privacy-preserving language model with homomorphic encryption (2), THOR: Secure transformer inference with homomorphic encryption (2), SWIFT: Shallow and SIMD-Aware CKKS Functional Bootstrapping for Low-Latency (1), Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption (1) …