一句话：该教程从软件工程视角综述基础模型驱动应用（FMware）的最新研究与工业实践。
基础模型（如 GPT-4）催生了 FMware，即由基础模型驱动的应用，构成新一代软件并在新范式下开发。尽管 FM 的随机性带来诸多挑战，FMware 已在软件工程研究（如测试生成）和工业产品（如 GitHub Copilot）中广泛采用。本教程将介绍工程 FMware 的最新研究与工业实践。其视角扎根于软件工程而非人工智能，以确保 FSE 参与者能够理解。教程不报告单一实证评估，而是梳理领域现状、挑战与实践，为构建和演化 FMware 提供软件工程指导。
Foundation Models (FMs) like GPT-4 have given rise to FMware, FM-powered applications representing a new generation of software that is developed under new paradigms. FMware has been widely adopted in both software engineering (SE) research (e.g., test generation) and industrial products (e.g., GitHub copilot), despite the numerous challenges introduced by the stochastic nature of FMs. In our tutorial, we will present the latest research and industry practices in engineering FMware. Our tutorial's perspective is firmly rooted in SE rather than artificial intelligence (AI), ensuring its accessibility to FSE participants.
@inproceedings{paperbot9162,
title = {A Tutorial on Software Engineering for FMware},
author = {Filipe Cogo and Gopi Krishnan Rajbahadur and Dayi Lin and Keheliya Gallaba and Benjamin Rombaut and Gustavo Oliva and Jiahuei (Justina) Lin and Kirill Vasilevski and Ahmed E. Hassan},
booktitle = {FSE},
year = {2025},
doi = {10.1145/3696630.3728621}
}