一句话：提出 EasyFHE，一个类 PyTorch 的 GPU 全同态加密编程框架，通过高层接口与 lowering 优化遍简化开发并提升性能。
在医疗、金融等敏感领域，机器学习越来越多地使用全同态加密（FHE）保护用户数据与模型。FHE 的内在并行性天然契合 GPU，但仅优化 GPU kernel 不足以支撑端到端 FHE 应用开发：FHE 方案本身复杂、GPU 细节繁琐，且高内存需求、细粒度内存操作与冗余计算带来优化难题。本文提出 EasyFHE 框架，提供类 PyTorch 的高层接口以定义 FHE 计算逻辑，自动处理实现选择、内存管理等底层任务；并设计优化框架，在将高层 FHE 程序 lowering 为 GPU kernel 的过程中应用定制优化遍，系统性消除性能瓶颈。相比 SOTA 开源 GPU FHE 库，EasyFHE 独特支持内存需求超出典型 GPU 容量的 FHE 程序，平均加速 2.88×，峰值达 4.39×。
In sensitive domains such as healthcare and finance, machine learning increasingly employs Fully Homomorphic Encryption (FHE) to secure both user data and models. Although FHE's intrinsic parallelism naturally aligns with GPU architectures, optimizing GPU kernels alone remains insufficient for efficient end-to-end FHE application development. The inherent complexity of FHE schemes and intricate GPU-specific details impede developers from focusing on high-level program logic. Additionally, FHE's high memory requirements, fine-grained memory operations, and redundant computations introduce further optimization challenges, resulting in inefficiencies even when GPU kernels are individually optimized. This paper introduces EasyFHE, a framework designed to simplify the development and optimization of GPU-accelerated FHE applications. Similar to PyTorch, EasyFHE provides high-level interfaces for defining computational logic while automatically handling low-level tasks, such as implementation selection and memory management. Furthermore, it incorporates an optimization framework that systematically addresses performance bottlenecks by applying tailored optimization passes during the lowering from high-level FHE programs to GPU kernels. Compared to state-of-the-art open-source GPU FHE libraries, EasyFHE uniquely supports FHE programs with memory requirements exceeding typical GPU capacities, achieving an average speedup of 2.88× with a peak of 4.39×.
@inproceedings{0004WF0Y0J26,
title = {A Framework for Developing and Optimizing Fully Homomorphic Encryption Programs on GPUs},
author = {Jianyu Zhao and Xueyu Wu and Guang Fan and Mingzhe Zhang and Shoumeng Yan and Lei Ju and Zhuoran Ji},
booktitle = {ASPLOS},
year = {2026},
doi = {10.1145/3779212.3790120}
}