# SENTRY: Statistical Reliability Analysis of Vision Transformers Under Soft Errors

arXiv: https://arxiv.org/abs/2606.07620 · HTML: https://mlsanity.com/paper/2606.07620/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 24 references that are in the map.
- First posted on arXiv 30 May 2026.

## Builds on (most cited first, 8 of 8)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)
- [Soft Error Reliability Analysis of Vision Transformers](https://mlsanity.com/paper/2302.10468/) (2023)
- [Noise-Tolerance GPU-based Age Estimation Using ResNet-50](https://mlsanity.com/paper/2305.00848/) (2023)
- [RESQ: A Unified Framework for REliability- and Security Enhancement of Quantized Deep Neural Networks](https://mlsanity.com/paper/2603.15413/) (2026)
- [Mix-and-Match Pruning: Globally Guided Layer-Wise Sparsification of DNNs](https://mlsanity.com/paper/2603.20280/) (2026)
- [PhD Thesis Summary: Methods for Reliability Assessment and Enhancement of Deep Neural Network Hardware Accelerators](https://mlsanity.com/paper/2603.08724/) (2026)

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

- [WARD: Runtime Workload-Adaptive Vision TRansformer Framework for Dependable Edge AI](https://mlsanity.com/paper/2609.17556/) (2026)

## Abstract (arXiv)

With the growth of Vision Transformers in safety-critical domains like autonomous systems and medical imaging, ensuring their reliability against soft errors is paramount. While ViTs offer state-of-the-art accuracy, their massive parameter counts render exhaustive fault injection campaigns infeasible. To bridge this gap, a statistical fault injection framework is presented, leveraging finite-population sampling theory to provide formal reliability guarantees. It is demonstrated that failure rates are bounded within a 1% margin at 99% confidence using only a few thousand samples, regardless of model scale. This methodology achieves up to a 10,700 times reduction in experimental cost compared to exhaustive approaches, while preserving the ability to localize vulnerabilities across architectural components. Through extensive evaluation of different architectures like ViT-Tiny and ViT-Small, a highly non-uniform reliability landscape is uncovered. It is shown that while only 3% of FP32 bit-flips result in failure, the vast majority of these events lead to catastrophic accuracy collapse. Specific vulnerabilities are localized to normalization layers and critical exponent bits within the IEEE-754 format, providing a mathematical foundation and actionable insights for the design of hardened, edge-deployed ViT architectures.