# GPU-Fuzz: Finding Memory Errors in Deep Learning Frameworks

arXiv: https://arxiv.org/abs/2602.10478 · HTML: https://mlsanity.com/paper/2602.10478/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 29 references that are in the map.
- First posted on arXiv 11 Feb 2026.

## Builds on (most cited first, 7 of 7)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [cuDNN: Efficient Primitives for Deep Learning](https://mlsanity.com/paper/1410.0759/) (2014)
- [A guide to convolution arithmetic for deep learning](https://mlsanity.com/paper/1603.07285/) (2016)
- [NNSmith: Generating Diverse and Valid Test Cases for Deep Learning Compilers](https://mlsanity.com/paper/2207.13066/) (2022)
- [Fuzzing Deep Learning Compilers with HirGen](https://mlsanity.com/paper/2208.02193/) (2022)
- [Guardian: Safe GPU Sharing in Multi-Tenant Environments](https://mlsanity.com/paper/2401.09290/) (2024)

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

- [The Correctness Illusion in LLM-Generated GPU Kernels](https://mlsanity.com/paper/2606.20128/) (2026)

## Abstract (arXiv)

GPU memory errors are a critical threat to deep learning (DL) frameworks, leading to crashes or even security issues. We introduce GPU-Fuzz, a fuzzer locating these issues efficiently by modeling operator parameters as formal constraints. GPU-Fuzz utilizes a constraint solver to generate test cases that systematically probe error-prone boundary conditions in GPU kernels. Applied to PyTorch, TensorFlow, and PaddlePaddle, we uncovered 13 unknown bugs, demonstrating the effectiveness of GPU-Fuzz in finding memory errors.