# ENFOR-SA: End-to-end Cross-layer Transient Fault Injector for Efficient and Accurate DNN Reliability Assessment on Systolic Arrays

arXiv: https://arxiv.org/abs/2602.00909 · HTML: https://mlsanity.com/paper/2602.00909/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jan 2026.

## Built on by (most cited first, 1 of 1)

- [TreeFI: Value-Aware Statistical Fault Injection for Deep Neural Networks](https://mlsanity.com/paper/2609.04912/) (2026)

## Abstract (arXiv)

Recent advances in deep learning have produced highly accurate but increasingly large and complex DNNs, making traditional fault-injection techniques impractical. Accurate fault analysis requires RTL-accurate hardware models. However, this significantly slows evaluation compared with software-only approaches, particularly when combined with expensive HDL instrumentation. In this work, we show that such high-overhead methods are unnecessary for systolic array (SA) architectures and propose ENFOR-SA, an end-to-end framework for DNN transient fault analysis on SAs. Our two-step approach employs cross-layer simulation and uses RTL SA components only during fault injection, with the rest executed at the software level. Experiments on CNNs and Vision Transformers demonstrate that ENFOR-SA achieves RTL-accurate fault injection with only 6% average slowdown compared to software-based injection, while delivering at least two orders of magnitude speedup (average 569×) over full-SoC RTL simulation and a 2.03× improvement over a state-of-the-art cross-layer RTL injection tool. ENFOR-SA code is publicly available at https://github.com/rafaabt/ENFOR-SA.