# Analysis of LLM Vulnerability to GPU Soft Errors: An Instruction-Level Fault Injection Study

arXiv: https://arxiv.org/abs/2601.19912 · HTML: https://mlsanity.com/paper/2601.19912/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 48 references that are in the map.
- First posted on arXiv 25 Dec 2025.

## Builds on (most cited first, 16 of 16)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Understanding Silent Data Corruption in LLM Training](https://mlsanity.com/paper/2502.12340/) (2025)
- [Soft Error Reliability Analysis of Vision Transformers](https://mlsanity.com/paper/2302.10468/) (2023)
- [Fault-Tolerant Deep Learning: A Hierarchical Perspective](https://mlsanity.com/paper/2204.01942/) (2022)
- [ReaLM: Reliable and Efficient Large Language Model Inference with Statistical Algorithm-Based Fault Tolerance](https://mlsanity.com/paper/2503.24053/) (2025)
- [MRFI: An Open Source Multi-Resolution Fault Injection Framework for Neural Network Processing](https://mlsanity.com/paper/2306.11758/) (2023)
- [Statistical Modeling of Soft Error Influence on Neural Networks](https://mlsanity.com/paper/2210.05876/) (2022)
- [Understanding the Effects of Permanent Faults in GPU's Parallelism Management and Control Units](https://mlsanity.com/paper/2306.10856/) (2023)
- [Evaluating Different Fault Injection Abstractions on the Assessment of DNN SW Hardening Strategies](https://mlsanity.com/paper/2412.08466/) (2024)

## Abstract (arXiv)

Large language models (LLMs) are highly compute- and memory-intensive, posing significant demands on high-performance GPUs. At the same time, advances in GPU technology driven by shrinking transistor sizes and lower operating voltages have made these devices increasingly susceptible to soft errors. While prior work has examined GPU reliability, most studies have focused on general-purpose applications or conventional neural networks mostly used for vision tasks such as classification and detection. In contrast, systematic analysis of modern large-scale LLMs remains limited, despite their rapid adoption in diverse application scenarios. Given the unique characteristics of LLMs, their resilience to soft errors may differ substantially from earlier models. To bridge this gap, we conduct the first instruction-level fault injection study of LLM inference. Our approach reveals reliability characteristics from multiple perspectives, highlighting the effects of model architecture, parameter scale, and task complexity. These findings provide new insights into LLM reliability and inform the design of more effective fault tolerance mechanisms.