Browse by task¶
Declared in extracted details¶
Values reported by the structured extraction (field tasks).
- image classification (29): Converting transformers to polynomial form for secure inference over homomorphic encryption; ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour; CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption; Efficient Multi-Key Homomorphic Encryption with Packed Ciphertexts with Application to Oblivious Neural Network Inference; Programmable Bootstrapping Enables Efficient Homomorphic Inference of Deep Neural Networks; GAZELLE: A Low Latency Framework for Secure Neural Network Inference; Delphi: A Cryptographic Inference Service for Neural Networks; CrypTFlow2: Practical 2-Party Secure Inference; Cheetah: Lean and Fast Secure Two-Party Deep Neural Network Inference; CrypTen: Secure Multi-Party Computation Meets Machine Learning; Orion: A Fully Homomorphic Encryption Framework for Deep Learning; DaCapo: Automatic Bootstrapping Management for Efficient Fully Homomorphic Encryption; TenSEAL: A Library for Encrypted Tensor Operations Using Homomorphic Encryption; Federated Learning: An approach with Hybrid Homomorphic Encryption; BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption; FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System; Towards the AlexNet Moment for Homomorphic Encryption: HCNN, the First Homomorphic CNN on Encrypted Data With GPUs; Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption; HHEML: Hybrid Homomorphic Encryption for Privacy-Preserving Machine Learning on Edge; Swift: Fast Secure Neural Network Inference With Fully Homomorphic Encryption; Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption; Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network; DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption; SpyKing—Privacy-preserving framework for Spiking Neural Networks; Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning; Homomorphic Encryption for Machine Learning and Artificial Intelligence Applications; On Fully Homomorphic Encryption for Privacy-Preserving Deep Learning; Towards Case-Optimized Hybrid Homomorphic Encryption; [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md)
- private transformer inference (22): THOR: Secure transformer inference with homomorphic encryption; ARION: Attention-optimized transformer inference on encrypted data; ELLMo: Packing- and Depth-Aware Encrypted Transformer Inference; TRICYCLE: Private transformer inference with tricyclic encodings; Encryption-friendly LLM architecture; Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity; STIP: Efficient and secure non-interactive transformer inference via compact packing; Converting transformers to polynomial form for secure inference over homomorphic encryption; THEMIS: A Co-Designed System for Encrypted Transformer Inference; Terrazzo: Memory-Aware GPU Framework for Private Transformer Inference; Zephyr: GPU-Efficient Homomorphic Encryption for Privacy-Preserving Transformer Inference; Error Propagation-Aware Scale Design for Efficient Homomorphic Encryption-Based LLM Inference; ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour; CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption; EncFormer: Secure and Efficient Transformer Inference over Encrypted Data; THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption; MPCFormer: Fast, Performant and Private Transformer Inference with MPC; PUMA: Secure Inference of LLaMA-7B in Five Minutes; BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers; BumbleBee: Secure Two-party Inference Framework for Large Transformers; Centaur: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference; SHAFT: Secure, Handy, Accurate, and Fast Transformer Inference
- text classification (14): ARION: Attention-optimized transformer inference on encrypted data; Encryption-friendly LLM architecture; Powerformer: Efficient and high-accuracy privacy-preserving language model with homomorphic encryption; MOAI: Module-optimizing architecture for non-interactive secure transformer inference; STIP: Efficient and secure non-interactive transformer inference via compact packing; ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour; MPCFormer: Fast, Performant and Private Transformer Inference with MPC; PUMA: Secure Inference of LLaMA-7B in Five Minutes; BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers; Ditto: Quantization-aware Secure Inference of Transformers upon MPC; CrypTen: Secure Multi-Party Computation Meets Machine Learning; Privft: Private and fast text classification with homomorphic encryption; Improving language understanding by generative pretraining; [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md)
- CKKS bootstrapping (12): A scalable multi-GPU framework for encrypted large-model inference; Improved Bootstrapping for Approximate Homomorphic Encryption; Efficient Bootstrapping for Approximate Homomorphic Encryption with Non-Sparse Keys; High-Precision Bootstrapping of RNS-CKKS Homomorphic Encryption Using Optimal Minimax Polynomial Approximation and Inverse Sine Function; Better Bootstrapping for Approximate Homomorphic Encryption; Plaintext-Ciphertext Matrix Multiplication and FHE Bootstrapping: Fast and Fused; Grafting: Decoupled Scale Factors and Modulus in RNS-CKKS; SHIP: A Shallow and Highly Parallelizable CKKS Bootstrapping Algorithm; PaCo: Bootstrapping for CKKS via Partial CoeffToSlot; Over 100x Faster Bootstrapping in Fully Homomorphic Encryption through Memory-centric Optimization with GPUs; BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption; CKKS Bootstrapping in less than 10ms
- language modeling (12): An Open-Source End-to-End FHE Implementation for Privacy-Preserving Llama 3 8B Inference; HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation; CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption; Ditto: Quantization-aware Secure Inference of Transformers upon MPC; FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System; Improving language understanding by generative pretraining; Llama: Open and efficient foundation language models; Pointer sentinel mixture models; The LAMBADA dataset: Word prediction requiring a broad discourse context; fairseq: A fast, extensible toolkit for sequence modeling; Bitnet: Scaling 1-bit transformers for large language models; GPT-4 System Card
- federated learning (11): An Efficient and Privacy-Preserving Federated Learning Approach Based on Homomorphic Encryption; Federated Learning: An approach with Hybrid Homomorphic Encryption; Federated Learning with Quantum Computing and Fully Homomorphic Encryption: A Novel Computing Paradigm Shift in Privacy-Preserving ML; BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption; Secure Federated Learning With Fully Homomorphic Encryption for IoT Communications; FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System; Secure federated learning applied to medical imaging with fully homomorphic encryption; An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption; FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users; PPFLHE: A privacy-preserving federated learning scheme with homomorphic encryption for healthcare data; Efficiency Optimization Techniques in Privacy-Preserving Federated Learning With Homomorphic Encryption: A Brief Survey
- FHE bootstrapping (10): Bootstrapping Bits with CKKS; Bootstrapping for Approximate Homomorphic Encryption; Efficient Bootstrapping in Fully Homomorphic Encryption for Matrix Arithmetic; Low-Depth Bootstrapping for Matrix-Native FHE; Bootstrapping for Approximate Homomorphic Encryption with Negligible Failure-Probability by Using Sparse-Secret Encapsulation; Bootstrapping for BGV and BFV Revisited; Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping; Bootstrapping GBFV with CKKS; Faster bootstrapping with polynomial error; LFHE: Fully Homomorphic Encryption with Bootstrapping Key Size Less than a Megabyte
- private LLM inference (10): SoK: Private LLM Inference using Approximate Homomorphic Encryption; EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption; Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache; TRICYCLE: Private transformer inference with tricyclic encodings; ENSI: Efficient non-interactive secure inference for large language models; Practical secure inference algorithm for a fine-tuned large language model based on fully homomorphic encryption; An Open-Source End-to-End FHE Implementation for Privacy-Preserving Llama 3 8B Inference; CipherGPT: Secure Two-Party GPT Inference; Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability; Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions
- private information retrieval (8): Multiparty Homomorphic Encryption from Ring-Learning-with-Errors; HEonGPU: a GPU-based Fully Homomorphic Encryption Library 1.0; Fully Homomorphic Encryption for Machine Learning; Efficient Fully Homomorphic Encryption from (Standard) LWE; Evaluating 2-DNF Formulas on Ciphertexts; Breaking the Circuit Size Barrier for Secure Computation Under DDH; One server for the price of two: Simple and fast single-server private information retrieval; SPIRAL: Fast, high-rate single-server PIR via FHE composition
- bootstrapping (7): FHEW: Bootstrapping Homomorphic Encryption in Less Than a Second; TFHE: Fast Fully Homomorphic Encryption over the Torus; F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption; ARK: Fully Homomorphic Encryption Accelerator with Runtime Data Generation and Inter-Operation Key Reuse; BASALISC: Programmable Hardware Accelerator for BGV Fully Homomorphic Encryption; Fully Homomorphic Encryption for Cyclotomic Prime Moduli; FHE for SIMD Arithmetic Logic Units with Amortized O(1) Bootstrapping per Ciphertext
- private neural network inference (7): Programmable Bootstrapping Enables Efficient Homomorphic Inference of Deep Neural Networks; Orion: A Fully Homomorphic Encryption Framework for Deep Learning; F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption; BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption; Swift: Fast Secure Neural Network Inference With Fully Homomorphic Encryption; A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption; Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities
- binary classification (6): Private pathological assessment via machine learning and homomorphic encryption; Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach; Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption; Evaluation of Privacy-aware Support Vector Machine (SVM) Learning using Homomorphic Encryption; Machine Learning Training on Encrypted Data with TFHE; Private svm inference on encrypted data
- private CNN inference (6): TenSEAL: A Library for Encrypted Tensor Operations Using Homomorphic Encryption; Towards the AlexNet Moment for Homomorphic Encryption: HCNN, the First Homomorphic CNN on Encrypted Data With GPUs; Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption; Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption; Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network; Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives
- LLM inference (4): MOAI: Module-optimizing architecture for non-interactive secure transformer inference; PUMA: Secure Inference of LLaMA-7B in Five Minutes; Confidential Computing on NVIDIA Hopper GPUs: A Performance Benchmark Study; PrivateLoRA for Efficient Privacy Preserving LLM
- natural language inference (4): THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption; BERT: Pre-training of deep bidirectional transformers for language understanding; Improving language understanding by generative pretraining; GLUE: A multi-task benchmark and analysis platform for natural language understanding
- NLP (4): Iron: Private Inference on Transformers; Centaur: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference; SHAFT: Secure, Handy, Accurate, and Fast Transformer Inference; ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs
- private inference (4): CrypTen: Secure Multi-Party Computation Meets Machine Learning; A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference; Non-interactive Fully Encrypted Machine Learning Protocol for Inference; Recent advances of privacy-preserving machine learning based on (Fully) Homomorphic Encryption
- encrypted logistic regression training (3): A Full RNS Variant of Approximate Homomorphic Encryption; Efficient Homomorphic Evaluation on Large Intervals; F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption
- encrypted matrix multiplication (3): Fully Homomorphic Encryption for Matrix Arithmetic; Trace-Factored BigSwitch for Matrix-Friendly FHE; Generalized BGV, BFV, and CKKS for Homomorphic Encryption over Matrix Rings
- logistic regression training (3): CHIMERA: Combining Ring-LWE-based Fully Homomorphic Encryption Schemes; Over 100x Faster Bootstrapping in Fully Homomorphic Encryption through Memory-centric Optimization with GPUs; Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption
- machine translation (3): MPC-Minimized Secure LLM Inference; Attention is all you need; fairseq: A fast, extensible toolkit for sequence modeling
- natural language understanding (3): PrivateLoRA for Efficient Privacy Preserving LLM; BERT: Pre-training of deep bidirectional transformers for language understanding; GLUE: A multi-task benchmark and analysis platform for natural language understanding
- privacy-preserving machine learning (3): Recent advances of privacy-preserving machine learning based on (Fully) Homomorphic Encryption; A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption; Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning
- private Transformer inference (3): Secure transformer inference made non-interactive; Iron: Private Inference on Transformers; SecFormer: Fast and Accurate Privacy-Preserving Inference for Transformer Models via SMPC
- secure neural network inference (3): GAZELLE: A Low Latency Framework for Secure Neural Network Inference; Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference; 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
- sentiment analysis (3): THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption; Integrating large language models in financial investments and market analysis: A survey; Privft: Private and fast text classification with homomorphic encryption
- threshold decryption (3): Threshold Cryptosystems From Threshold Fully Homomorphic Encryption; Low Communication Threshold FHE from Standard (Module-)LWE; Ajax: Fast Threshold Fully Homomorphic Encryption without Noise Flooding
- benchmarking (2): Phantom: A CUDA-Accelerated Word-Wise Homomorphic Encryption Library; Pasta: A Case for Hybrid Homomorphic Encryption
- classification (2): Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption; Non-interactive Fully Encrypted Machine Learning Protocol for Inference
- descriptive statistics (2): Simple Functional Encryption Schemes for Inner Products; Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption
- electronic voting (2): A Generalisation, a Simplification and Some Applications of Paillier's Probabilistic Public-Key System; Evaluating 2-DNF Formulas on Ciphertexts
- encrypted CNN inference (2): A scalable multi-GPU framework for encrypted large-model inference; HeLayers: A Tile Tensors Framework for Large Neural Networks on Encrypted Data
- encrypted database lookup (2): F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption; Fully Homomorphic SIMD Operations
- FHE acceleration (2): ARK: Fully Homomorphic Encryption Accelerator with Runtime Data Generation and Inter-Operation Key Reuse; BASALISC: Programmable Hardware Accelerator for BGV Fully Homomorphic Encryption
- FHE hardware acceleration (2): BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption; TREBUCHET: Fully Homomorphic Encryption Accelerator for Deep Computation
- functional bootstrapping (2): The Power of Low Rank: Fast CKKS Functional Bootstrapping for High-Precision Lookup Tables; Large-Plaintext Functional Bootstrapping in FHE with Small Bootstrapping Keys
- genome-wide association studies (2): CHIMERA: Combining Ring-LWE-based Fully Homomorphic Encryption Schemes; Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics
- GLUE text classification (2): ELLMo: Packing- and Depth-Aware Encrypted Transformer Inference; Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity
- homomorphic comparison (2): Grafting: Decoupled Scale Factors and Modulus in RNS-CKKS; Faster Logical Operations from Discrete CKKS
- Machine Learning as a Service (2): A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption; A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption
- medical image classification (2): Private pathological assessment via machine learning and homomorphic encryption; A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference
- paraphrase detection (2): THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption; GLUE: A multi-task benchmark and analysis platform for natural language understanding
- privacy-preserving ML inference (2): HE-MAN – Homomorphically Encrypted MAchine learning with oNnx models; Approximate homomorphic encryption based privacy-preserving machine learning: a survey
- privacy-preserving model aggregation (2): An Efficient and Privacy-Preserving Federated Learning Approach Based on Homomorphic Encryption; FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users
- privacy-preserving SVM inference (2): Evaluation of Privacy-aware Support Vector Machine (SVM) Learning using Homomorphic Encryption; Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption
- privacy-preserving training (2): Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption; ABY3: A Mixed Protocol Framework for Machine Learning
- private LLM fine-tuning (2): Encryption-friendly LLM architecture; Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions
- private training (2): CrypTen: Secure Multi-Party Computation Meets Machine Learning; Recent advances of privacy-preserving machine learning based on (Fully) Homomorphic Encryption
- private transformer/LLM inference (2): SIGMA: Secure GPT Inference with Function Secret Sharing; SoK: Private transformer-based model inference
- question answering (2): BERT: Pre-training of deep bidirectional transformers for language understanding; Improving language understanding by generative pretraining
- secure model aggregation (2): Federated Learning: An approach with Hybrid Homomorphic Encryption; FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System
- secure transformer inference (2): MOAI: Module-optimizing architecture for non-interactive secure transformer inference; Nimbus: Secure and Efficient Two-Party Inference for Transformers
- secure Transformer inference (2): Ditto: Quantization-aware Secure Inference of Transformers upon MPC; CRAfT: Constant-Round Non-Polynomial Approximation and P2P Network for Secure Transformer Inference
- sentiment classification (2): GLUE: A multi-task benchmark and analysis platform for natural language understanding; Recursive deep models for semantic compositionality over a sentiment treebank
- text classification (GLUE) (2): THOR: Secure transformer inference with homomorphic encryption; THEMIS: A Co-Designed System for Encrypted Transformer Inference
- text generation (2): CipherGPT: Secure Two-Party GPT Inference; InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model
- transciphering (2): The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption; Towards Case-Optimized Hybrid Homomorphic Encryption
- transciphering / hybrid homomorphic encryption (2): XBOOT: Free-XOR Gates for CKKS with Applications to Transciphering; Transciphering Framework for Approximate Homomorphic Encryption
- 64-bit unsigned integer arithmetic (1): Efficient Homomorphic Integer Computer from CKKS
- abstractive summarization (1): fairseq: A fast, extensible toolkit for sequence modeling
- activation functions (ReLU, sigmoid) (1): TFHE Gets Real: An Efficient and Flexible Homomorphic Floating-Point Arithmetic
- activation functions (ReLU, sigmoid, tanh) (1): SWIFT: Shallow and SIMD-Aware CKKS Functional Bootstrapping for Low-Latency
- additively homomorphic encryption (1): Linearly Homomorphic Encryption from DDH
- anti-money laundering detection (1): Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering
- approximate arithmetic on encrypted data (1): Bootstrapping for Approximate Homomorphic Encryption
- approximate arithmetic on encrypted real/complex numbers (1): Homomorphic Encryption for Arithmetic of Approximate Numbers
- approximate arithmetic over real numbers (1): A Full RNS Variant of Approximate Homomorphic Encryption
- approximate homomorphic arithmetic (1): Better Bootstrapping for Approximate Homomorphic Encryption
- approximate real/complex homomorphic arithmetic (1): Approximate Homomorphic Encryption with Reduced Approximation Error
- arbitrary lookup-table evaluation (1): General Functional Bootstrapping using CKKS
- arbitrary-precision homomorphic arithmetic (1): Grafting: Decoupled Scale Factors and Modulus in RNS-CKKS
- arithmetic circuit evaluation (1): Multiparty Computation from Somewhat Homomorphic Encryption
- arithmetic-to-Boolean conversion (1): High-Precision Homomorphic ALU over Arbitrary Moduli with O (1) Bootstrapping
- attack on instance encoding (1): Neuracrypt is not private
- author profiling (1): Beyond memorization: Violating privacy via inference with large language models
- automatic speech recognition (1): Theory, analysis, and best practices for sigmoid selfattention
- autoregressive decoding (1): Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability
- autoregressive language modeling (1): Theory, analysis, and best practices for sigmoid selfattention
- autoregressive text generation (1): Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache
- Beaver multiplication triple generation (1): Multiparty Homomorphic Encryption from Ring-Learning-with-Errors
- bias and toxicity evaluation (1): Llama: Open and efficient foundation language models
- binary sentiment classification (1): Recursive deep models for semantic compositionality over a sentiment treebank
- binary text classification (1): Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data
- binary tree multiplication (1): Revisiting Homomorphic Encryption Schemes for Finite Fields
- biometric authentication (1): Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption
- biometric recognition (1): Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication
- bit-wise operations on encrypted integers (1): Faster Logical Operations from Discrete CKKS
- boolean circuit evaluation (1): FHEW: Bootstrapping Homomorphic Encryption in Less Than a Second
- Boolean circuit evaluation (1): TFHE: Fast Fully Homomorphic Encryption over the Torus
- Boolean circuit evaluation in CKKS (1): XBOOT: Free-XOR Gates for CKKS with Applications to Transciphering
- boolean gate bootstrapping (1): FINAL: Faster FHE Instantiated with NTRU and LWE
- branching program evaluation (1): GG-GSW: Chosen-Ciphertext Secure Leveled FHE From Gadget Trapdoors
- breast cancer detection (masses and microcalcifications) in mammograms (1): Secure federated learning applied to medical imaging with fully homomorphic encryption
- Byzantine-robust aggregation (1): FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users
- Certificate Transparency auditing (1): One server for the price of two: Simple and fast single-server private information retrieval
- certified robustness via randomized smoothing (1): Private and reliable neural network inference
- chatbot (1): MPC-Minimized Secure LLM Inference
- chosen-ciphertext secure homomorphic evaluation (1): GG-GSW: Chosen-Ciphertext Secure Leveled FHE From Gadget Trapdoors
- circuit privacy / model confidentiality (1): Fully Homomorphic Encryption for Machine Learning
- CKKS arithmetic bootstrapping (1): Stay in Your Lane: Fast Arithmetic over Large Finite Rings with CKKS
- CKKS bootstrapping and functional bootstrapping (1): High-Precision Functional Bootstrapping for CKKS from Fourier Extension
- client-server outsourced computation (1): The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption
- clinical decision support (1): Large language models in medicine
- clinical documentation and administration (1): Large language models in medicine
- cloud computing security (1): A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption
- cloud data processing (1): A comprehensive survey on secure outsourced computation and its applications
- cloud machine learning (1): Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities
- cloud-assisted secure multiparty computation (1): On-the-Fly Multiparty Computation on the Cloud via Multikey Fully Homomorphic Encryption
- code generation (1): MPC-Minimized Secure LLM Inference
- collaborative training on split data (1): Machine Learning Training on Encrypted Data with TFHE
- common sense reasoning (1): PrivateLoRA for Efficient Privacy Preserving LLM
- commonsense natural language inference (1): HellaSwag: Can a machine really finish your sentence?
- confidential blockchain / smart contracts (discussed) (1): The Beginner's Textbook for Fully Homomorphic Encryption
- confidential computing (1): Confidential Computing on NVIDIA Hopper GPUs: A Performance Benchmark Study
- constituency parsing (1): Attention is all you need
- Conway's Game of Life (1): BLEACH: Cleaning Errors in Discrete Computations over CKKS
- cryptanalysis of FHE-friendly symmetric ciphers (1): Cryptanalysis of Symmetric Primitives over Rings and a Key Recovery Attack on Rubato
- cryptographic neural network inference (1): Delphi: A Cryptographic Inference Service for Neural Networks
- data movement over packed ciphertexts (1): Algorithms in HElib
- defense against data poisoning (1): FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users
- defense against gradient inversion and model poisoning attacks (1): BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption
- dialogue (1): fairseq: A fast, extensible toolkit for sequence modeling
- distributed (threshold) decryption (1): Revisiting Shamir Secret Sharing for Threshold Fully Homomorphic Encryption
- document classification (1): Privft: Private and fast text classification with homomorphic encryption
- document retrieval (1): [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md)
- e-voting (1): Benaloh's Dense Probabilistic Encryption Revisited
- e-voting (motivating application) (1): Linearly Homomorphic Encryption from DDH
- edge AI (1): An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption
- edge AI inference (1): Large language model inference acceleration: A comprehensive hardware perspective
- Edge-AI analytics (1): Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning
- efficient attention (1): [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md)
- efficient client-side encryption of real-valued data for CKKS (1): Transciphering Framework for Approximate Homomorphic Encryption
- enabling IND-CPA-D security, threshold FHE and circuit privacy via noise flooding (1): Leveraging Discrete CKKS to Bootstrap in High Precision
- encrypted attention (transformer inference kernels) (1): Trace-Factored BigSwitch for Matrix-Friendly FHE
- encrypted binary circuit evaluation (1): BLEACH: Cleaning Errors in Discrete Computations over CKKS
- encrypted CNN inference (handwritten digit classification) (1): Secure outsourced matrix computation and application to neural networks
- encrypted database queries (discussed) (1): The Beginner's Textbook for Fully Homomorphic Encryption
- encrypted deep learning inference as a service (1): Fully Homomorphically Encrypted Deep Learning as a Service
- encrypted deep-learning inference (1): Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning
- encrypted LLM inference (prefill) (1): A scalable multi-GPU framework for encrypted large-model inference
- encrypted machine learning inference (motivating application) (1): Generalized BGV, BFV, and CKKS for Homomorphic Encryption over Matrix Rings
- encrypted matrix arithmetic (1): Efficient Bootstrapping in Fully Homomorphic Encryption for Matrix Arithmetic
- encrypted matrix arithmetic for privacy-preserving ML inference (1): Low-Depth Bootstrapping for Matrix-Native FHE
- encrypted ML training (1): Machine Learning Training on Encrypted Data with TFHE
- encrypted spiking neural network inference (1): SpyKing—Privacy-preserving framework for Spiking Neural Networks
- encrypted tensor operations (1): TenSEAL: A Library for Encrypted Tensor Operations Using Homomorphic Encryption
- encrypted text classification (1): SoK: Private LLM Inference using Approximate Homomorphic Encryption
- encrypted token generation (1): SoK: Private LLM Inference using Approximate Homomorphic Encryption
- evaluation key compression / client-server communication reduction (1): LFHE: Fully Homomorphic Encryption with Bootstrapping Key Size Less than a Megabyte
- evaluation of boolean circuits / binary polynomials on encrypted data (1): Fully Homomorphic Encryption with Relatively Small Key and Ciphertext Sizes
- evaluation of boolean circuits on encrypted bits (1): Fully Homomorphic Encryption over the Integers
- evaluation of multiplicative inverse, exponential, logistic function and discrete Fourier transform (1): Homomorphic Encryption for Arithmetic of Approximate Numbers
- exact HE arithmetic over finite fields (1): Revisiting Homomorphic Encryption Schemes for Finite Fields
- exact modular arithmetic on encrypted data (1): Bootstrapping for BGV and BFV Revisited
- exact modular arithmetic over large finite rings and fields (1): Stay in Your Lane: Fast Arithmetic over Large Finite Rings with CKKS
- extractive question answering (1): SQuAD: 100,000+ questions for machine comprehension of text
- face image classification (1): Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption
- face recognition (1): HE-MAN – Homomorphically Encrypted MAchine learning with oNnx models
- fairness certification (1): Private and reliable neural network inference
- fall detection (1): Secure human action recognition by encrypted neural network inference
- few-shot benchmark evaluation (1): Llama: Open and efficient foundation language models
- FHE acceleration (NTT, homomorphic multiplication, bootstrapping) (1): TensorFHE: Achieving Practical Computation on Encrypted Data Using GPGPU
- FHE bootstrapping benchmarking (1): Demystifying bootstrapping in fully homomorphic encryption
- FHE parameter selection (1): Security guidelines for implementing homomorphic encryption
- FHE program compilation / bootstrapping placement (1): DaCapo: Automatic Bootstrapping Management for Efficient Fully Homomorphic Encryption
- FHE scheme switching (1): CHIMERA: Combining Ring-LWE-based Fully Homomorphic Encryption Schemes
- FHE secure against active (chosen-ciphertext) attacks (1): Fully Homomorphic Encryption with Chosen-Ciphertext Security from LWE
- FHE standardization (1): Security guidelines for implementing homomorphic encryption
- FHEW bootstrapping (1): Efficient FHEW Bootstrapping with Small Evaluation Keys, and Applications to Threshold Homomorphic Encryption
- financial forecasting (1): Integrating large language models in financial investments and market analysis: A survey
- fine-grained 5-class sentiment (1): Recursive deep models for semantic compositionality over a sentiment treebank
- fingerprint authentication (1): Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication
- fraud detection (1): Private svm inference on encrypted data
- fully homomorphic encryption (1): Somewhat Practical Fully Homomorphic Encryption
- fully homomorphic encryption construction (1): Homomorphic Encryption from Learning with Errors: Conceptually-Simpler, Asymptotically-Faster, Attribute-Based
- functional bootstrapping / look-up table evaluation (1): SWIFT: Shallow and SIMD-Aware CKKS Functional Bootstrapping for Low-Latency
- gadget decomposition / external product (1): Accelerating HE Operations from Key Decomposition Technique
- general homomorphic evaluation / outsourced computation (1): Fully Homomorphic Encryption without Modulus Switching from Classical GapSVP
- general secure computation (1): MP-SPDZ: A Versatile Framework for Multi-Party Computation
- general secure multiparty computation (1): Two Round Multiparty Computation via Multi-Key FHE
- generative LLM inference (1): Large language model inference acceleration: A comprehensive hardware perspective
- genomic algorithms (paternity and ancestry testing) (1): Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics
- GPT inference (1): SIGMA: Secure GPT Inference with Function Secret Sharing
- GPU acceleration of FHE (1): Cheddar: A Swift Fully Homomorphic Encryption Library Designed for GPU Architectures
- GPU acceleration of HE primitives (1): Phantom: A CUDA-Accelerated Word-Wise Homomorphic Encryption Library
- GPU memory expansion / tensor offloading (1): Cost-efficient llm training with lifetimeaware tensor offloading via gpudirect storage
- grammatical error correction (1): fairseq: A fast, extensible toolkit for sequence modeling
- graph-based transaction classification (1): Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering
- handwritten digit classification (1): HE-MAN – Homomorphically Encrypted MAchine learning with oNnx models
- healthcare analytics (1): Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach
- healthcare data analytics (1): Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption
- healthcare image classification (1): PPFLHE: A privacy-preserving federated learning scheme with homomorphic encryption for healthcare data
- healthcare ML (motivation) (1): Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning
- healthcare NLP (1): Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data
- heart disease prediction (1): Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach
- high-precision CKKS bootstrapping (1): Leveraging Discrete CKKS to Bootstrap in High Precision
- high-precision exact integer arithmetic on encrypted data (1): High-Precision Exact FHE Made Simple, General, and Fast
- high-precision homomorphic integer arithmetic (P-384, Curve25519, RSA-1024, RSA-2048 moduli) (1): High-Precision Homomorphic ALU over Arbitrary Moduli with O (1) Bootstrapping
- high-precision integer and rational arithmetic on encrypted data (1): High-Precision Arithmetic in Homomorphic Encryption
- high-precision integer arithmetic (1): Fully Homomorphic Encryption for Cyclotomic Prime Moduli
- high-precision lookup table evaluation (1): The Power of Low Rank: Fast CKKS Functional Bootstrapping for High-Precision Lookup Tables
- homomorphic 2D convolution (1): HeLayers: A Tile Tensors Framework for Large Neural Networks on Encrypted Data
- homomorphic AES evaluation (1): Fully Homomorphic SIMD Operations
- homomorphic binary circuit evaluation (1): Bootstrapping Bits with CKKS
- homomorphic bootstrapping on GPU (1): Terrazzo: Memory-Aware GPU Framework for Private Transformer Inference
- homomorphic ciphertext compression / transciphering (1): Stream ciphers: A Practical Solution for Efficient Homomorphic-Ciphertext Compression
- homomorphic comparison and bit shift (1): Efficient Homomorphic Integer Computer from CKKS
- homomorphic encryption primitive benchmarking (1): Improved Security for a Ring-Based Fully Homomorphic Encryption Scheme
- homomorphic evaluation of boolean (NAND) circuits (1): Ring packing and amortized FHEW bootstrapping
- homomorphic evaluation of GeLU, Sigmoid, exp, floor and S-box lookup tables (1): High-Precision Functional Bootstrapping for CKKS from Fourier Extension
- homomorphic evaluation of symmetric ciphers (1): Stream ciphers: A Practical Solution for Efficient Homomorphic-Ciphertext Compression
- homomorphic floating-point arithmetic (1): TFHE Gets Real: An Efficient and Flexible Homomorphic Floating-Point Arithmetic
- homomorphic integer arithmetic and bitwise operations on machine words (1): FHE for SIMD Arithmetic Logic Units with Amortized O(1) Bootstrapping per Ciphertext
- homomorphic key switching (1): Accelerating HE Operations from Key Decomposition Technique
- homomorphic linear algebra (1): Algorithms in HElib
- homomorphic look-up table evaluation (1): Large-Plaintext Functional Bootstrapping in FHE with Small Bootstrapping Keys
- homomorphic ML inference (1): HEonGPU: a GPU-based Fully Homomorphic Encryption Library 1.0
- homomorphic neural network inference (1): Fully Homomorphic Encryption for Machine Learning
- homomorphic Softmax evaluation (1): Fast and Accurate Homomorphic Softmax Evaluation
- human action recognition (1): Secure human action recognition by encrypted neural network inference
- human activity recognition (1): An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption
- hybrid homomorphic encryption (1): HEonGPU: a GPU-based Fully Homomorphic Encryption Library 1.0
- hybrid homomorphic encryption / transciphering (1): Pasta: A Case for Hybrid Homomorphic Encryption
- identity-based and attribute-based homomorphic encryption (1): Homomorphic Encryption from Learning with Errors: Conceptually-Simpler, Asymptotically-Faster, Attribute-Based
- image processing (1): EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation
- IND-CPA^D security of approximate homomorphic encryption (1): Securing Approximate Homomorphic Encryption Using Differential Privacy
- inference on LoRA fine-tuned domain models (1): Practical secure inference algorithm for a fine-tuned large language model based on fully homomorphic encryption
- inner product (1): Revisiting Homomorphic Encryption Schemes for Finite Fields
- inner product benchmark (1): MP-SPDZ: A Versatile Framework for Multi-Party Computation
- inner-product computation on encrypted vectors (1): Simple Functional Encryption Schemes for Inner Products
- inverse and inverse square root evaluation (1): SWIFT: Shallow and SIMD-Aware CKKS Functional Bootstrapping for Low-Latency
- IoT security / attack detection in smart cities (1): Secure Federated Learning With Fully Homomorphic Encryption for IoT Communications
- key recovery attack (1): On the Security of Homomorphic Encryption on Approximate Numbers
- language model benchmark evaluation (1): EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption
- language modeling (next token prediction) (1): Converting transformers to polynomial form for secure inference over homomorphic encryption
- language modeling (perplexity) (1): ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation
- language modelling (1): PUMA: Secure Inference of LLaMA-7B in Five Minutes
- language understanding (1): Nimbus: Secure and Efficient Two-Party Inference for Transformers
- large-precision integer arithmetic on encrypted data (1): Bootstrapping GBFV with CKKS
- linguistic acceptability (1): GLUE: A multi-task benchmark and analysis platform for natural language understanding
- LLM auditing (1): zkLLM: Zero Knowledge Proofs for Large Language Models
- LLM legal advice (1): (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice
- LLM personalization (1): PrivateLoRA for Efficient Privacy Preserving LLM
- LLM privacy attack (1): Beyond memorization: Violating privacy via inference with large language models
- LLM quantization (1): Bitnet: Scaling 1-bit transformers for large language models
- LLM training (1): Cost-efficient llm training with lifetimeaware tensor offloading via gpudirect storage
- logistic and exponential function approximation (1): On the Security of Homomorphic Encryption on Approximate Numbers
- logistic regression inference (1): ModHE: Modular Homomorphic Encryption Using Module Lattices: Potentials and Limitations
- logistic regression training (workload modeling) (1): TREBUCHET: Fully Homomorphic Encryption Accelerator for Deep Computation
- long-sequence classification (1): [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md)
- look-up table evaluation (1): TFHE: Fast Fully Homomorphic Encryption over the Torus
- machine learning as a service (1): Private svm inference on encrypted data
- machine learning inference on encrypted data (1): Building Privacy-Preserving AI system: AI Inference on Encrypted Data with Fully Homomorphic Encryption
- medical data classification (1): Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics
- medical diagnosis (1): Private pathological assessment via machine learning and homomorphic encryption
- medical education (1): Large language models in medicine
- medical PII detection (1): Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data
- medical question answering (1): Large language models in medicine
- mental poker (1): Probabilistic Encryption
- milk yield forecasting (sequence regression) (1): Fully Homomorphically Encrypted Deep Learning as a Service
- mixed integer / fixed-point computation (1): BLEACH: Cleaning Errors in Discrete Computations over CKKS
- ML-as-a-service prediction APIs (1): Stealing machine learning models via prediction APIs
- MLaaS (1): SoK: Privacy-preserving computation techniques for deep learning
- MLaaS cloud inference (motivation) (1): FIDESlib: A Fully-Fledged Open-Source FHE Library for Efficient CKKS on GPUs
- model extraction / model stealing (1): Stealing machine learning models via prediction APIs
- MPC-friendly symmetric encryption (1): Rasta: A cipher with low ANDdepth and few ANDs per bit
- multi-client private inference (1): CRAfT: Constant-Round Non-Polynomial Approximation and P2P Network for Secure Transformer Inference
- multi-key homomorphic encryption (1): Revisiting Module Lattice-based Homomorphic Encryption and Application to Secure-MPC
- multi-party computation on data encrypted under multiple keys (1): FINALLY: A Multi-Key FHE Scheme Based on NTRU and LWE
- multi-party computation over encrypted data (1): Ajax: Fast Threshold Fully Homomorphic Encryption without Noise Flooding
- multi-party FHE (1): Low Communication Threshold FHE from Standard (Module-)LWE
- multi-party trust computation (1): Benaloh's Dense Probabilistic Encryption Revisited
- multi-precision sign and floor evaluation (1): General Functional Bootstrapping using CKKS
- multilingual sentence embeddings (1): fairseq: A fast, extensible toolkit for sequence modeling
- multiple-choice question answering (1): Think you have solved question answering? try ARC, the AI2 reasoning challenge
- multiple-choice sentence completion (1): HellaSwag: Can a machine really finish your sentence?
- named entity recognition (1): THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption
- natural language understanding (GLUE tasks) (1): SecFormer: Fast and Accurate Privacy-Preserving Inference for Transformer Models via SMPC
- neural network inference compilation (1): EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation
- noise analysis and parameter selection (1): On the precision loss in approximate homomorphic encryption
- non-interactive verifiable secret sharing (1): Benaloh's Dense Probabilistic Encryption Revisited
- non-linear function evaluation (Step, ReLU) (1): The Power of Low Rank: Fast CKKS Functional Bootstrapping for High-Precision Lookup Tables
- non-polynomial function evaluation (1): PEGASUS: Bridging Polynomial and Non-polynomial Evaluations in Homomorphic Encryption
- object detection (1): Orion: A Fully Homomorphic Encryption Framework for Deep Learning
- oblivious neural network inference (1): Efficient Multi-Key Homomorphic Encryption with Packed Ciphertexts with Application to Oblivious Neural Network Inference
- oblivious permutation (1): Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping
- outsourced ML computation (1): Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware
- parameter-efficient fine-tuning (1): PrivateLoRA for Efficient Privacy Preserving LLM
- paraphrase / pair similarity (1): Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy
- paraphrase and sentiment classification (1): BERT: Pre-training of deep bidirectional transformers for language understanding
- paraphrase identification (1): Automatically constructing a corpus of sentential paraphrases
- pathological assessment (1): Private pathological assessment via machine learning and homomorphic encryption
- personal attribute inference from text (1): Beyond memorization: Violating privacy via inference with large language models
- plaintext-ciphertext matrix multiplication (1): Plaintext-Ciphertext Matrix Multiplication and FHE Bootstrapping: Fast and Fused
- polynomial approximation of modular reduction (1): High-Precision Bootstrapping of RNS-CKKS Homomorphic Encryption Using Optimal Minimax Polynomial Approximation and Inverse Sine Function
- polynomial approximation of non-polynomial functions (1): Efficient Homomorphic Evaluation on Large Intervals
- polynomial evaluation (1): Revisiting Homomorphic Encryption Schemes for Finite Fields
- portfolio optimization (1): Integrating large language models in financial investments and market analysis: A survey
- privacy-preserving approximate computing (1): FIDESlib: A Fully-Fledged Open-Source FHE Library for Efficient CKKS on GPUs
- privacy-preserving deep learning inference (1): On Fully Homomorphic Encryption for Privacy-Preserving Deep Learning
- privacy-preserving deep learning training and inference (1): SoK: Privacy-preserving computation techniques for deep learning
- privacy-preserving home monitoring / telemedicine (1): Secure human action recognition by encrypted neural network inference
- privacy-preserving inference (1): ABY3: A Mixed Protocol Framework for Machine Learning
- privacy-preserving LLM attention (motivation) (1): Fully Homomorphic Encryption for Matrix Arithmetic
- privacy-preserving machine learning (discussed as application) (1): The Beginner's Textbook for Fully Homomorphic Encryption
- privacy-preserving ML inference on edge/IoT devices (1): HHEML: Hybrid Homomorphic Encryption for Privacy-Preserving Machine Learning on Edge
- privacy-preserving ML training (1): Approximate homomorphic encryption based privacy-preserving machine learning: a survey
- privacy-preserving ML training and inference (1): SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes
- privacy-preserving MLaaS on third-party cloud (1): Building Privacy-Preserving AI system: AI Inference on Encrypted Data with Fully Homomorphic Encryption
- privacy-preserving model training on encrypted data (1): Practical Considerations of Fully Homomorphic Encryption in Privacy-Preserving Machine Learning
- privacy-preserving neural network training (1): SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training
- privacy-preserving neural network training (image classification) (1): SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine Learning
- privacy-preserving numerical simulation of PDEs (linear advection, finite differences) (1): Secure numerical simulations using fully homomorphic encryption
- privacy-preserving training on medical images (1): Neuracrypt is not private
- private classification in outsourced computation (1): Private machine learning classification based on fully homomorphic encryption
- private decision tree evaluation (1): PEGASUS: Bridging Polynomial and Non-polynomial Evaluations in Homomorphic Encryption
- private DNN inference (1): Cheetah: Lean and Fast Secure Two-Party Deep Neural Network Inference
- private DNN inference with FHE (1): DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption
- private image classification inference (1): Activate Me!: Designing Efficient Activation Functions for Privacy-Preserving Machine Learning with Fully Homomorphic Encryption
- private language model inference (1): Powerformer: Efficient and high-accuracy privacy-preserving language model with homomorphic encryption
- private LLM data selection (1): Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions
- private LLM embedding-as-a-service inference (1): Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy
- private LLM inference (attention) (1): Fast and Accurate Homomorphic Softmax Evaluation
- private LLM inference (autoregressive decoding) (1): ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation
- private LLM inference (encrypted autoregressive decoding) (1): HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation
- private LLM inference (prefill/summarization and token generation) (1): 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)
- private LLM inference / prompt privacy (1): InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model
- private LLM token generation (1): BumbleBee: Secure Two-party Inference Framework for Large Transformers
- private logistic regression (1): BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption
- private NN inference (1): Private and reliable neural network inference
- private set intersection (1): Multiparty Homomorphic Encryption from Ring-Learning-with-Errors
- private text generation (1): SoK: Private transformer-based model inference
- private Transformer/LLM inference (1): Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives
- private transformer/LLM inference (motivation) (1): Plaintext-Ciphertext Matrix Multiplication and FHE Bootstrapping: Fast and Fused
- PSI with computation (1): Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping
- public-key encryption (1): Probabilistic Encryption
- quantum machine learning (1): Federated Learning with Quantum Computing and Fully Homomorphic Encryption: A Novel Computing Paradigm Shift in Privacy-Preserving ML
- reactor structural health monitoring and nondestructive evaluation (motivating application) (1): Homomorphic Encryption for Machine Learning and Artificial Intelligence Applications
- reading comprehension (1): SQuAD: 100,000+ questions for machine comprehension of text
- real-number homomorphic arithmetic (1): Approximate Homomorphic Encryption over the Conjugate-invariant Ring
- reasoning (1): Think you have solved question answering? try ARC, the AI2 reasoning challenge
- red teaming (1): GPT-4 System Card
- reproducibility assessment (1): SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training
- research writing (1): Large language models in medicine
- responsible AI policy design (1): (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice
- retrieval-augmented generation in TEEs (1): Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability
- risk assessment (1): Integrating large language models in financial investments and market analysis: A survey
- S-box evaluation (1): General Functional Bootstrapping using CKKS
- safety evaluation (1): GPT-4 System Card
- scheme conversion (1): Faster Logical Operations from Discrete CKKS
- science question answering (1): Think you have solved question answering? try ARC, the AI2 reasoning challenge
- secure biomedical image analysis with transformers (1): SoK: Private transformer-based model inference
- secure cloud-based biometric matching (1): Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication
- secure computation among multiple data providers (1): Efficient Multi-Key Homomorphic Encryption with Packed Ciphertexts with Application to Oblivious Neural Network Inference
- secure K-means clustering (1): PEGASUS: Bridging Polynomial and Non-polynomial Evaluations in Homomorphic Encryption
- secure LLM inference (1): MPC-Minimized Secure LLM Inference
- secure machine learning (as application domain) (1): Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping
- secure multi-party computation (1): Revisiting Module Lattice-based Homomorphic Encryption and Application to Secure-MPC
- secure outsourced computation (1): A comprehensive survey on secure outsourced computation and its applications
- secure outsourced machine learning (1): A comprehensive survey on secure outsourced computation and its applications
- secure outsourced matrix computation (1): Secure outsourced matrix computation and application to neural networks
- secure search and update of distributed data (1): Function Secret Sharing
- secure two-party computation (1): Breaking the Circuit Size Barrier for Secure Computation Under DDH
- secure two-party DNN inference (1): CrypTFlow2: Practical 2-Party Secure Inference
- security parameter recommendation for homomorphic encryption (1): Homomorphic encryption standard
- self-supervised image representation learning (SimCLR, BYOL, MAE) (1): Theory, analysis, and best practices for sigmoid selfattention
- semantic compositionality (1): Recursive deep models for semantic compositionality over a sentiment treebank
- semantic similarity (1): THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption
- sentence classification (1): Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy
- sentence similarity (1): Improving language understanding by generative pretraining
- sentence similarity regression (1): BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers
- SIMD FHE arithmetic over small prime plaintext moduli (1): Fully Homomorphic Encryption on the Ring of Gaussian Periods
- spam detection (1): Privft: Private and fast text classification with homomorphic encryption
- speech recognition (1): CrypTen: Secure Multi-Party Computation Meets Machine Learning
- statistical analysis over encrypted data (1): A Full RNS Variant of Approximate Homomorphic Encryption
- stock selection (1): Integrating large language models in financial investments and market analysis: A survey
- story generation (1): fairseq: A fast, extensible toolkit for sequence modeling
- supervised image classification (1): Theory, analysis, and best practices for sigmoid selfattention
- survival analysis (1): Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption
- text entailment classification (RTE) (1): Zephyr: GPU-Efficient Homomorphic Encryption for Privacy-Preserving Transformer Inference
- text understanding (1): The LAMBADA dataset: Word prediction requiring a broad discourse context
- textual entailment (1): Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy
- textual similarity (1): GLUE: A multi-task benchmark and analysis platform for natural language understanding
- threshold FHE (1): Revisiting Shamir Secret Sharing for Threshold Fully Homomorphic Encryption
- threshold homomorphic encryption (1): Efficient FHEW Bootstrapping with Small Evaluation Keys, and Applications to Threshold Homomorphic Encryption
- threshold key generation (1): Ajax: Fast Threshold Fully Homomorphic Encryption without Noise Flooding
- threshold PRFs and key generation (1): Threshold Cryptosystems From Threshold Fully Homomorphic Encryption
- threshold signatures (1): Threshold Cryptosystems From Threshold Fully Homomorphic Encryption
- token generation (1): PUMA: Secure Inference of LLaMA-7B in Five Minutes
- topic classification (1): Privft: Private and fast text classification with homomorphic encryption
- trading (1): Integrating large language models in financial investments and market analysis: A survey
- training models on encrypted data (1): Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics
- transciphering / FHE ciphertext-expansion removal (1): Rasta: A cipher with low ANDdepth and few ANDs per bit
- transciphering for approximate FHE (1): Cryptanalysis of Symmetric Primitives over Rings and a Key Recovery Attack on Rubato
- transciphering symmetric ciphertexts into CKKS ciphertexts (1): Rubato : Noisy Ciphers for Approximate Homomorphic Encryption (Full Version)
- transformer inference (1): ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs
- two-party privacy-preserving machine learning (1): Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives
- variance computation (1): On the Security of Homomorphic Encryption on Approximate Numbers
- verifiable and private DNN inference (1): Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware
- verifiable computation (1): Evaluating 2-DNF Formulas on Ciphertexts
- verifiable LLM inference (1): zkLLM: Zero Knowledge Proofs for Large Language Models
- weighted automata and arithmetic in leveled mode (1): TFHE: Fast Fully Homomorphic Encryption over the Torus
- word prediction (1): The LAMBADA dataset: Word prediction requiring a broad discourse context
- zero-knowledge proof of attention (1): zkLLM: Zero Knowledge Proofs for Large Language Models
- zero-knowledge proof system arithmetic (motivation) (1): Fully Homomorphic Encryption for Cyclotomic Prime Moduli
- zero-shot and few-shot downstream evaluation (1): Bitnet: Scaling 1-bit transformers for large language models
Mentions in full text¶
Papers ordered by how often the term occurs (count in brackets). Only papers with a local PDF are counted.
Language modeling / generation (130 papers)¶
Large language model inference acceleration: A comprehensive hardware perspective (122), The Beginner's Textbook for Fully Homomorphic Encryption (95), Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache (43), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (42), Fully Homomorphic Encryption on the Ring of Gaussian Periods (34), PrivateLoRA for Efficient Privacy Preserving LLM (28), MPC-Minimized Secure LLM Inference (25), Pointer sentinel mixture models (21), Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability (18), Efficient Bootstrapping in Fully Homomorphic Encryption for Matrix Arithmetic (18), Function Secret Sharing (15), ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation (15), fairseq: A fast, extensible toolkit for sequence modeling (13), FHE for SIMD Arithmetic Logic Units with Amortized O(1) Bootstrapping per Ciphertext (13), SoK: Private LLM Inference using Approximate Homomorphic Encryption (13), High-Precision Exact FHE Made Simple, General, and Fast (12), On the precision loss in approximate homomorphic encryption (11), 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) (11), Theory, analysis, and best practices for sigmoid selfattention (10), Secure numerical simulations using fully homomorphic encryption (10), Llama: Open and efficient foundation language models (10), SPIRAL: Fast, high-rate single-server PIR via FHE composition (9), Efficient Bootstrapping for Approximate Homomorphic Encryption with Non-Sparse Keys (9), General Functional Bootstrapping using CKKS (9), Homomorphic Encryption for Arithmetic of Approximate Numbers (9), EncFormer: Secure and Efficient Transformer Inference over Encrypted Data (9), HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation (9), Improving language understanding by generative pretraining (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), On the Security of Homomorphic Encryption on Approximate Numbers (8), Stay in Your Lane: Fast Arithmetic over Large Finite Rings with CKKS (7), BERT: Pre-training of deep bidirectional transformers for language understanding (6), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (6), Probabilistic Encryption (6), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (6), Improved Bootstrapping for Approximate Homomorphic Encryption (6), Approximate Homomorphic Encryption with Reduced Approximation Error (6), Secure outsourced matrix computation and application to neural networks (5), Bitnet: Scaling 1-bit transformers for large language models (5), Security guidelines for implementing homomorphic encryption (5), SoK: Private transformer-based model inference (5), Large language models in medicine (5), Breaking the Circuit Size Barrier for Secure Computation Under DDH (5), Relaxed Functional Bootstrapping: A New Perspective on BGV/BFV Bootstrapping (5), Confidential Computing on NVIDIA Hopper GPUs: A Performance Benchmark Study (5), High-Precision Arithmetic in Homomorphic Encryption (5), Better Bootstrapping for Approximate Homomorphic Encryption (5), The LAMBADA dataset: Word prediction requiring a broad discourse context (4), Attention is all you need (4), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (4), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (4), Generalized BGV, BFV, and CKKS for Homomorphic Encryption over Matrix Rings (4), ModHE: Modular Homomorphic Encryption Using Module Lattices: Potentials and Limitations (4), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (4), Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption (4), Phantom: A CUDA-Accelerated Word-Wise Homomorphic Encryption Library (4), Trace-Factored BigSwitch for Matrix-Friendly FHE (4), Programmable Bootstrapping Enables Efficient Homomorphic Inference of Deep Neural Networks (4), Fully Homomorphic Encryption for Matrix Arithmetic (4), SHIP: A Shallow and Highly Parallelizable CKKS Bootstrapping Algorithm (4) …
Medical / healthcare (123 papers)¶
A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (128), Large language models in medicine (125), Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption (90), Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach (81), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (63), Private pathological assessment via machine learning and homomorphic encryption (52), PPFLHE: A privacy-preserving federated learning scheme with homomorphic encryption for healthcare data (40), A comprehensive survey on secure outsourced computation and its applications (39), Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption (28), Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data (26), Secure federated learning applied to medical imaging with fully homomorphic encryption (24), Private and reliable neural network inference (19), (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice (19), Secure Federated Learning With Fully Homomorphic Encryption for IoT Communications (19), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (17), GPT-4 System Card (13), Secure human action recognition by encrypted neural network inference (13), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (12), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (11), Secure numerical simulations using fully homomorphic encryption (10), Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities (10), Efficiency Optimization Techniques in Privacy-Preserving Federated Learning With Homomorphic Encryption: A Brief Survey (10), Federated Learning with Quantum Computing and Fully Homomorphic Encryption: A Novel Computing Paradigm Shift in Privacy-Preserving ML (10), Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication (9), Private machine learning classification based on fully homomorphic encryption (8), An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption (8), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (8), Neuracrypt is not private (7), Fully Homomorphic Encryption for Machine Learning (7), 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 (7), Non-interactive Fully Encrypted Machine Learning Protocol for Inference (7), An Efficient and Privacy-Preserving Federated Learning Approach Based on Homomorphic Encryption (7), BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers (7), GAZELLE: A Low Latency Framework for Secure Neural Network Inference (7), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (6), SoK: Privacy-preserving computation techniques for deep learning (6), Private svm inference on encrypted data (6), DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption (6), High-Precision Arithmetic in Homomorphic Encryption (6), Homomorphic Encryption for Arithmetic of Approximate Numbers (6), EncFormer: Secure and Efficient Transformer Inference over Encrypted Data (6), Privft: Private and fast text classification with homomorphic encryption (5), ABY3: A Mixed Protocol Framework for Machine Learning (5), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (5), Multiparty Homomorphic Encryption from Ring-Learning-with-Errors (5), ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour (5), Secure outsourced matrix computation and application to neural networks (4), Stealing machine learning models via prediction APIs (4), Llama: Open and efficient foundation language models (4), A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (4), Activate Me!: Designing Efficient Activation Functions for Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (4), Better Bootstrapping for Approximate Homomorphic Encryption (4), Bootstrapping for Approximate Homomorphic Encryption (4), On the Security of Homomorphic Encryption on Approximate Numbers (4), Homomorphic encryption standard (3), SoK: Private transformer-based model inference (3), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (3), Transciphering Framework for Approximate Homomorphic Encryption (3), ModHE: Modular Homomorphic Encryption Using Module Lattices: Potentials and Limitations (3), Practical Considerations of Fully Homomorphic Encryption in Privacy-Preserving Machine Learning (3) …
Finance (96 papers)¶
Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering (37), Integrating large language models in financial investments and market analysis: A survey (36), Beyond memorization: Violating privacy via inference with large language models (11), Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities (10), (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice (9), 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 (7), Secure transformer inference made non-interactive (7), Private and reliable neural network inference (6), Secure numerical simulations using fully homomorphic encryption (6), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (6), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (6), CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption (6), Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption (5), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (4), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (4), Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy (4), A comprehensive survey on secure outsourced computation and its applications (3), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (3), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (3), Private svm inference on encrypted data (3), The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption (3), Transciphering Framework for Approximate Homomorphic Encryption (3), Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption (3), Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication (3), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (3), Pointer sentinel mixture models (2), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (2), DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption (2), HHEML: Hybrid Homomorphic Encryption for Privacy-Preserving Machine Learning on Edge (2), PPFLHE: A privacy-preserving federated learning scheme with homomorphic encryption for healthcare data (2), An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption (2), FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System (2), Federated Learning with Quantum Computing and Fully Homomorphic Encryption: A Novel Computing Paradigm Shift in Privacy-Preserving ML (2), Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data (2), Bootstrapping for Approximate Homomorphic Encryption (2), Multiparty Homomorphic Encryption from Ring-Learning-with-Errors (2), A Full RNS Variant of Approximate Homomorphic Encryption (2), EncFormer: Secure and Efficient Transformer Inference over Encrypted Data (2), Terrazzo: Memory-Aware GPU Framework for Private Transformer Inference (2), STIP: Efficient and secure non-interactive transformer inference via compact packing (2), Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity (2), ELLMo: Packing- and Depth-Aware Encrypted Transformer Inference (2), Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache (2), SoK: Private LLM Inference using Approximate Homomorphic Encryption (2), Homomorphic encryption standard (1), One server for the price of two: Simple and fast single-server private information retrieval (1), SoK: Private transformer-based model inference (1), Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability (1), Privft: Private and fast text classification with homomorphic encryption (1), Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption (1), ABY3: A Mixed Protocol Framework for Machine Learning (1), A Generalisation, a Simplification and Some Applications of Paillier's Probabilistic Public-Key System (1), Rubato : Noisy Ciphers for Approximate Homomorphic Encryption (Full Version) (1), Multi-key Fully Homomorphic Encryption with Non-Interactive Setup in the Plain Model (1), Fully Homomorphic SIMD Operations (1), Revisiting Module Lattice-based Homomorphic Encryption and Application to Secure-MPC (1), Fully Homomorphic Encryption for Machine Learning (1), The Beginner's Textbook for Fully Homomorphic Encryption (1), A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (1), A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption (1) …
Genomics (61 papers)¶
Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (62), A comprehensive survey on secure outsourced computation and its applications (32), Collaborative privacy-preserving analysis of oncological data using multiparty homomorphic encryption (17), Private pathological assessment via machine learning and homomorphic encryption (10), Secure human action recognition by encrypted neural network inference (8), Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption (8), Secure numerical simulations using fully homomorphic encryption (7), Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities (7), Over 100x Faster Bootstrapping in Fully Homomorphic Encryption through Memory-centric Optimization with GPUs (7), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (6), Private svm inference on encrypted data (6), Recent advances of privacy-preserving machine learning based on (Fully) Homomorphic Encryption (6), Evaluation of Privacy-aware Support Vector Machine (SVM) Learning using Homomorphic Encryption (6), Approximate Homomorphic Encryption with Reduced Approximation Error (6), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (5), HEonGPU: a GPU-based Fully Homomorphic Encryption Library 1.0 (5), Secure outsourced matrix computation and application to neural networks (4), 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 (4), Multiparty Homomorphic Encryption from Ring-Learning-with-Errors (4), XBOOT: Free-XOR Gates for CKKS with Applications to Transciphering (3), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (3), Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach (3), CHIMERA: Combining Ring-LWE-based Fully Homomorphic Encryption Schemes (3), High-Precision Arithmetic in Homomorphic Encryption (3), Programmable Bootstrapping Enables Efficient Homomorphic Inference of Deep Neural Networks (3), Bootstrapping for Approximate Homomorphic Encryption (3), A Full RNS Variant of Approximate Homomorphic Encryption (3), Homomorphic Encryption for Arithmetic of Approximate Numbers (3), Automatically constructing a corpus of sentential paraphrases (2), Privft: Private and fast text classification with homomorphic encryption (2), Machine Learning Training on Encrypted Data with TFHE (2), A Survey on Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (2), Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network (2), Non-interactive Fully Encrypted Machine Learning Protocol for Inference (2), Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption (2), Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data (2), F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption (2), TensorFHE: Achieving Practical Computation on Encrypted Data Using GPGPU (2), OpenFHE: Open-Source Fully Homomorphic Encryption Library (2), Better Bootstrapping for Approximate Homomorphic Encryption (2), Efficient Multi-Key Homomorphic Encryption with Packed Ciphertexts with Application to Oblivious Neural Network Inference (2), Approximate Homomorphic Encryption over the Conjugate-invariant Ring (2), ENSI: Efficient non-interactive secure inference for large language models (2), ARION: Attention-optimized transformer inference on encrypted data (2), Homomorphic encryption standard (1), On the precision loss in approximate homomorphic encryption (1), Demystifying bootstrapping in fully homomorphic encryption (1), ABY3: A Mixed Protocol Framework for Machine Learning (1), Transciphering Framework for Approximate Homomorphic Encryption (1), Practical Considerations of Fully Homomorphic Encryption in Privacy-Preserving Machine Learning (1), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (1), Private machine learning classification based on fully homomorphic encryption (1), Efficiency Optimization Techniques in Privacy-Preserving Federated Learning With Homomorphic Encryption: A Brief Survey (1), BASALISC: Programmable Hardware Accelerator for BGV Fully Homomorphic Encryption (1), TREBUCHET: Fully Homomorphic Encryption Accelerator for Deep Computation (1), Revisiting Homomorphic Encryption Schemes for Finite Fields (1), Fast and Accurate Homomorphic Softmax Evaluation (1), Efficient Homomorphic Evaluation on Large Intervals (1), Accelerating HE Operations from Key Decomposition Technique (1), CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption (1) …
IoT / edge (37 papers)¶
Secure Federated Learning With Fully Homomorphic Encryption for IoT Communications (78), PrivateLoRA for Efficient Privacy Preserving LLM (41), An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption (12), Efficiency Optimization Techniques in Privacy-Preserving Federated Learning With Homomorphic Encryption: A Brief Survey (11), Federated Learning: An approach with Hybrid Homomorphic Encryption (9), An Efficient and Privacy-Preserving Federated Learning Approach Based on Homomorphic Encryption (8), HHEML: Hybrid Homomorphic Encryption for Privacy-Preserving Machine Learning on Edge (7), Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach (7), Large language model inference acceleration: A comprehensive hardware perspective (6), SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (4), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (4), LFHE: Fully Homomorphic Encryption with Bootstrapping Key Size Less than a Megabyte (4), PPFLHE: A privacy-preserving federated learning scheme with homomorphic encryption for healthcare data (4), Federated Learning with Quantum Computing and Fully Homomorphic Encryption: A Novel Computing Paradigm Shift in Privacy-Preserving ML (4), A comprehensive survey on secure outsourced computation and its applications (3), Exploring Bitslicing Architectures for Enabling FHE-Assisted Machine Learning (3), Fully Homomorphically Encrypted Deep Learning as a Service (3), FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System (3), BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption (3), The HHE Land: Exploring the Landscape of Hybrid Homomorphic Encryption (2), Secure human action recognition by encrypted neural network inference (2), Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication (2), Practical Considerations of Fully Homomorphic Encryption in Privacy-Preserving Machine Learning (1), Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference (1), Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network (1), Non-interactive Fully Encrypted Machine Learning Protocol for Inference (1), Evaluation of Privacy-Preserving Support Vector Machine (SVM) Learning Using Homomorphic Encryption (1), Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption (1), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (1), Application of Machine Learning-NLP Approach with Fully Homomorphic Encryption Techniques in Medical PII Data (1), BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption (1), F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption (1), Confidential Computing on NVIDIA Hopper GPUs: A Performance Benchmark Study (1), BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers (1), SIGMA: Secure GPT Inference with Function Secret Sharing (1), META-BTS: Bootstrapping Precision Beyond the Limit (1), Euston: Efficient and User-Friendly Secure Transformer Inference with Non-Interactivity (1)
Question answering (27 papers)¶
Improving language understanding by generative pretraining (14), Think you have solved question answering? try ARC, the AI2 reasoning challenge (7), SQuAD: 100,000+ questions for machine comprehension of text (7), Llama: Open and efficient foundation language models (7), BERT: Pre-training of deep bidirectional transformers for language understanding (6), Pointer sentinel mixture models (2), Iron: Private Inference on Transformers (2), SoK: Private LLM Inference using Approximate Homomorphic Encryption (2), GPT-4 System Card (1), Theory, analysis, and best practices for sigmoid selfattention (1), Attention is all you need (1), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (1), Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability (1), Integrating large language models in financial investments and market analysis: A survey (1), (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice (1), Large language models in medicine (1), Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy (1), Nimbus: Secure and Efficient Two-Party Inference for Transformers (1), BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers (1), CipherGPT: Secure Two-Party GPT Inference (1), PUMA: Secure Inference of LLaMA-7B in Five Minutes (1), THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption (1), ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation (1), Secure transformer inference made non-interactive (1), Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache (1), EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption (1), THOR: Secure transformer inference with homomorphic encryption (1)
Machine translation (24 papers)¶
fairseq: A fast, extensible toolkit for sequence modeling (13), Attention is all you need (12), Improving language understanding by generative pretraining (7), MPC-Minimized Secure LLM Inference (6), Automatically constructing a corpus of sentential paraphrases (5), Bitnet: Scaling 1-bit transformers for large language models (4), Llama: Open and efficient foundation language models (4), Pointer sentinel mixture models (3), BERT: Pre-training of deep bidirectional transformers for language understanding (3), ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs (2), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (2), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (2), (a)i am not a lawyer, but...: Engaging legal experts towards responsible llm policies for legal advice (2), Encryption-friendly LLM architecture (2), Theory, analysis, and best practices for sigmoid selfattention (1), The LAMBADA dataset: Word prediction requiring a broad discourse context (1), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (1), Privacypreserving llm inference in practice: A comparative survey of techniques, trade-offs, and deployability (1), A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption (1), DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption (1), InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model (1), MPCFormer: Fast, Performant and Private Transformer Inference with MPC (1), Iron: Private Inference on Transformers (1), Practical secure inference algorithm for a fine-tuned large language model based on fully homomorphic encryption (1)
Image classification (21 papers)¶
Approximate homomorphic encryption based privacy-preserving machine learning: a survey (3), Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption (3), Secure federated learning applied to medical imaging with fully homomorphic encryption (3), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (3), Converting transformers to polynomial form for secure inference over homomorphic encryption (3), Theory, analysis, and best practices for sigmoid selfattention (2), Private pathological assessment via machine learning and homomorphic encryption (2), Private and reliable neural network inference (1), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (1), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (1), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (1), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (1), Privacy-preserving neural networks with Homomorphic encryption: Challenges and opportunities (1), FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users (1), Federated Learning with Quantum Computing and Fully Homomorphic Encryption: A Novel Computing Paradigm Shift in Privacy-Preserving ML (1), SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine Learning (1), BumbleBee: Secure Two-party Inference Framework for Large Transformers (1), CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption (1), HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation (1), Secure transformer inference made non-interactive (1), ELLMo: Packing- and Depth-Aware Encrypted Transformer Inference (1)
Clustering (20 papers)¶
A comprehensive survey on secure outsourced computation and its applications (33), PEGASUS: Bridging Polynomial and Non-polynomial Evaluations in Homomorphic Encryption (22), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (14), An Efficient Architecture for Edge AI Federated Learning With Homomorphic Encryption (13), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (3), ABY3: A Mixed Protocol Framework for Machine Learning (3), Secure Federated Learning With Fully Homomorphic Encryption for IoT Communications (3), Exploring the future of privacy-preserving heart disease prediction: a fully homomorphic encryption-driven logistic regression approach (3), Centaur: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference (3), Touch of Privacy: A Homomorphic Encryption-Powered Deep Learning Framework for Fingerprint Authentication (2), On the Security of Homomorphic Encryption on Approximate Numbers (2), Large language model inference acceleration: A comprehensive hardware perspective (1), Towards efficient privacy-preserving machine learning: A systematic review from protocol, model, and system perspectives (1), Integrating large language models in financial investments and market analysis: A survey (1), Activate Me!: Designing Efficient Activation Functions for Privacy-Preserving Machine Learning with Fully Homomorphic Encryption (1), FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users (1), Federated Learning: An approach with Hybrid Homomorphic Encryption (1), MPC-Minimized Secure LLM Inference (1), BOLT: Privacy-Preserving, Accurate and Efficient Inference for Transformers (1), GAZELLE: A Low Latency Framework for Secure Neural Network Inference (1)
Text classification (18 papers)¶
Approximate homomorphic encryption based privacy-preserving machine learning: a survey (11), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (4), Fully Homomorphic Encryption with Applications to Privacy-Preserving Machine Learning (2), Private pathological assessment via machine learning and homomorphic encryption (2), Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy (2), Encryption-friendly LLM architecture (2), Automatically constructing a corpus of sentential paraphrases (1), Private svm inference on encrypted data (1), ModHE: Modular Homomorphic Encryption Using Module Lattices: Potentials and Limitations (1), Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption (1), Optimized Privacy-Preserving CNN Inference With Fully Homomorphic Encryption (1), BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption (1), Phantom: A CUDA-Accelerated Word-Wise Homomorphic Encryption Library (1), Centaur: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference (1), Fast and Accurate Homomorphic Softmax Evaluation (1), Converting transformers to polynomial form for secure inference over homomorphic encryption (1), THOR: Secure transformer inference with homomorphic encryption (1), SoK: Private LLM Inference using Approximate Homomorphic Encryption (1)
Sentiment analysis (15 papers)¶
Integrating large language models in financial investments and market analysis: A survey (12), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (9), Recursive deep models for semantic compositionality over a sentiment treebank (4), Split-and-Denoise: Protect Large Language Model Inference with Local Differential Privacy (4), Privft: Private and fast text classification with homomorphic encryption (3), [Skyformer: Remodel selfattention with gaussian kernel and nystr](../papers/skyformer-remodel-selfattention-with-gaussian-kernel-and-nystr.md) (1), Improving language understanding by generative pretraining (1), BERT: Pre-training of deep bidirectional transformers for language understanding (1), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (1), Cryptography-based privacy-preserving large language models: a lifecycle survey of frameworks, methods, and future directions (1), MPCFormer: Fast, Performant and Private Transformer Inference with MPC (1), THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption (1), ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour (1), ENSI: Efficient non-interactive secure inference for large language models (1), SoK: Private LLM Inference using Approximate Homomorphic Encryption (1)
Recommendation (13 papers)¶
SoK: Connecting the dots in privacy-preserving ML - systematization of MPC protocols and conversions between secret sharing schemes (2), GPT-4 System Card (1), ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs (1), A comprehensive survey on secure outsourced computation and its applications (1), Private svm inference on encrypted data (1), Privft: Private and fast text classification with homomorphic encryption (1), ABY3: A Mixed Protocol Framework for Machine Learning (1), SWIFT: Shallow and SIMD-Aware CKKS Functional Bootstrapping for Low-Latency (1), DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption (1), Towards the AlexNet Moment for Homomorphic Encryption: HCNN, the First Homomorphic CNN on Encrypted Data With GPUs (1), A privacy preserving machine learning framework for medical image analysis using quantized fully connected neural networks with TFHE based inference (1), BumbleBee: Secure Two-party Inference Framework for Large Transformers (1), THEMIS: A Co-Designed System for Encrypted Transformer Inference (1)
Speech recognition (10 papers)¶
Theory, analysis, and best practices for sigmoid selfattention (9), Fully Homomorphic Encryption for Machine Learning (4), CrypTen: Secure Multi-Party Computation Meets Machine Learning (4), A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption (3), Llama: Open and efficient foundation language models (2), Approximate homomorphic encryption based privacy-preserving machine learning: a survey (2), Recursive deep models for semantic compositionality over a sentiment treebank (1), Improving language understanding by generative pretraining (1), ABY3: A Mixed Protocol Framework for Machine Learning (1), DOReN: Toward Efficient Deep Convolutional Neural Networks with Fully Homomorphic Encryption (1)
Named entity recognition (8 papers)¶
BERT: Pre-training of deep bidirectional transformers for language understanding (4), THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption (3), The LAMBADA dataset: Word prediction requiring a broad discourse context (2), Pointer sentinel mixture models (2), SQuAD: 100,000+ questions for machine comprehension of text (1), Improving language understanding by generative pretraining (1), Integrating large language models in financial investments and market analysis: A survey (1), THOR: Secure transformer inference with homomorphic encryption (1)
Anomaly detection (3 papers)¶
Stealing machine learning models via prediction APIs (2), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (1), FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users (1)
Object detection (3 papers)¶
Orion: A Fully Homomorphic Encryption Framework for Deep Learning (8), SoK: Wildest dreams: Reproducible research in privacy-preserving neural network training (1), CRAfT: Constant-Round Non-Polynomial Approximation and P2P Network for Secure Transformer Inference (1)
Segmentation (2 papers)¶
Private and reliable neural network inference (1), Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics (1)