一句话：构建覆盖JAX、TensorFlow、PyTorch的112个可复现缺陷环境数据库，支撑深度学习库测试技术评估。
深度学习库（如PyTorch、TensorFlow）是当前AI应用的基础构件，但与其他软件一样可能存在缺陷。尽管已有大量缺陷发现技术被提出，该领域却缺乏经过整理的可复现缺陷集合，导致这些技术难以得到可信的评估。本文提出BugsInDLLs，一个经过整理的、可复现的深度学习库缺陷数据库，以填补这一空白。构建此类数据库面临独特挑战，例如为复现特定GPU相关缺陷，需要安装特定CUDA版本的驱动等。该数据集当前包含112个可复现环境，覆盖JAX、TensorFlow和PyTorch三个主流深度学习库。该工作为系统评估深度学习库测试技术提供了基础设施。
AI-enabled applications are prolific today. Deep Learning (DL) libraries, such as PyTorch and Tensorflow, provide the building blocks for the AI components of these applications. As any piece of software, these libraries can be buggy. An impressive number of bug-finding techniques to address this problem have been proposed, but the lack of a curated set of reproducible bugs in DL libraries hinders credible evaluation of these techniques. We present BugsInDLLs, a database of curated reproducible bugs to fill that gap. Unique challenges exist in this context, such as installing drivers of specific CUDA versions to reproduce certain GPU-related bugs. Our dataset currently consists of 112 environments to reproduce bugs across three popular DL libraries, namely, JAX, Tensorflow, and PyTorch.
DOI 原文 ·
@inproceedings{paperbot9602,
title = {BugsInDLLs : A Database of Reproducible Bugs in Deep Learning Libraries to Enable Systematic Evaluation of Testing Techniques},
author = {M M Abid Naziri and Aman Kumar Singh and Benjamin Wu and Feiran (Alex) Qin and Saikat Dutta and Marcelo d'Amorim},
booktitle = {ISSTA},
year = {2025},
doi = {10.1145/3713081.3731739}
}