Paper 2026/1502
Efficient Privacy-Preserving LSTM Inference on Encrypted Sequential Data
Abstract
Recent advances in fully homomorphic encryption (FHE) have enabled privacy-preserving machine learning directly over encrypted data. As a representative recurrent architecture, the long short-term memory (LSTM) network is widely used for modeling sequential dependencies, yet existing FHE-based LSTM inference schemes still suffer from high latency and limited scalability. In this paper, we present an efficient privacy-preserving LSTM inference protocol on encrypted sequential data based on FHE. We insert a lightweight normalization module before nonlinear activations to bound the hidden states, thereby enabling accurate low-degree polynomial approximations of the Sigmoid, Tanh, and inverse square-root functions via a hybrid Remez and least-squares strategy. We implement the proposed protocol using the Lattigo library, incorporating ciphertext packing optimization, rotation minimization, and SIMD-based parallelization. Experiments show that on standard text classification benchmarks, our encrypted LSTM achieves competitive accuracy compared with plaintext models and consistently outperforms the state-of-the-art FHE-based method, achieving up to a 4.7x speedup.
Metadata
- Available format(s)
-
PDF
- Category
- Applications
- Publication info
- Published elsewhere. Minor revision. Neurocomputing
- DOI
- 10.1016/j.neucom.2026.134508
- Keywords
- Privacy preserving machine learningHomomorphic encryptionLSTM
- Contact author(s)
-
heqiang @ cigit ac cn
chenjingwei @ cigit ac cn
wuwenyuan @ cigit ac cn
yongfeng @ cigit ac cn - History
- 2026-07-25: approved
- 2026-07-23: received
- See all versions
- Short URL
- https://ia.cr/2026/1502
- License
-
CC BY-NC-ND
BibTeX
@misc{cryptoeprint:2026/1502,
author = {Qiang He and Jingwei Chen and Wenyuan Wu and Yong Feng},
title = {Efficient Privacy-Preserving {LSTM} Inference on Encrypted Sequential Data},
howpublished = {Cryptology {ePrint} Archive, Paper 2026/1502},
year = {2026},
doi = {10.1016/j.neucom.2026.134508},
url = {https://eprint.iacr.org/2026/1502}
}
