FaRAccel: FPGA-Accelerated Defense Architecture for Efficient Bit-Flip Attack Resilience in Transformer Models

TL;DR

FaRAccel is a novel FPGA-based hardware architecture that accelerates and optimizes the FaR defense mechanism against Bit-Flip Attacks in Transformer models, significantly reducing latency and energy consumption.

Contribution

This work introduces the first hardware-accelerated implementation of FaR, enhancing its efficiency and practicality for real-world deployment on FPGA platforms.

Findings

Substantial reduction in inference latency.

Improved energy efficiency during defense operations.

Maintained robustness against Bit-Flip Attacks.

Abstract

Forget and Rewire (FaR) methodology has demonstrated strong resilience against Bit-Flip Attacks (BFAs) on Transformer-based models by obfuscating critical parameters through dynamic rewiring of linear layers. However, the application of FaR introduces non-negligible performance and memory overheads, primarily due to the runtime modification of activation pathways and the lack of hardware-level optimization. To overcome these limitations, we propose FaRAccel, a novel hardware accelerator architecture implemented on FPGA, specifically designed to offload and optimize FaR operations. FaRAccel integrates reconfigurable logic for dynamic activation rerouting, and lightweight storage of rewiring configurations, enabling low-latency inference with minimal energy overhead. We evaluate FaRAccel across a suite of Transformer models and demonstrate substantial reductions in FaR inference latency and improvement in energy efficiency, while maintaining the robustness gains of the original FaR methodology. To the best of our knowledge, this is the first hardware-accelerated defense against BFAs in Transformers, effectively bridging the gap between algorithmic resilience and efficient deployment on real-world AI platforms.