Breaking the Layer Barrier: Remodeling Private Transformer Inference with Hybrid CKKS and MPC

TL;DR

This paper introduces BLB, a novel framework that combines CKKS and MPC for private Transformer inference, significantly reducing communication and latency costs by fusing linear layers and enabling secure CKKS-MPC conversion.

Contribution

BLB breaks the layer barrier in private Transformer inference by fusing operators and introducing a secure CKKS-MPC conversion protocol, improving efficiency over existing methods.

Findings

Achieves 21x reduction in communication overhead compared to BOLT.

Reduces latency by up to 13x with GPU acceleration.

Supports efficient secure computation of fused linear operators.

Abstract

This paper presents an efficient framework for private Transformer inference that combines Homomorphic Encryption (HE) and Secure Multi-party Computation (MPC) to protect data privacy. Existing methods often leverage HE for linear layers (e.g., matrix multiplications) and MPC for non-linear layers (e.g., Softmax activation functions), but the conversion between HE and MPC introduces significant communication costs. The proposed framework, dubbed BLB, overcomes this by breaking down layers into fine-grained operators and further fusing adjacent linear operators, reducing the need for HE/MPC conversions. To manage the increased ciphertext bit width from the fused linear operators, BLB proposes the first secure conversion protocol between CKKS and MPC and enables CKKS-based computation of the fused operators. Additionally, BLB proposes an efficient matrix multiplication protocol for fused computation in Transformers. Extensive evaluations on BERT-base, BERT-large, and GPT2-base show that BLB achieves a $21\times$ reduction in communication overhead compared to BOLT (S\&P'24) and a $2\times$ reduction compared to Bumblebee (NDSS'25), along with latency reductions of $13\times$ and $1.8\times$, respectively, when leveraging GPU acceleration.