A Framework for Designing Foundation Model based Systems

05/09/2023
by   Qinghua Lu, et al.
0

The recent release of large language model (LLM) based chatbots, such as ChatGPT, has attracted significant attention on foundation models. It is widely believed that foundation models will serve as the fundamental building blocks for future AI systems. As foundation models are in their early stages, the design of foundation model based systems has not yet been systematically explored. There is little understanding about the impact of introducing foundation models in software architecture. Therefore, in this paper, we propose a taxonomy of foundation model based systems, which classifies and compares the characteristics of foundation models and design options of foundation model based systems. Our taxonomy comprises three categories: foundation model pretraining and fine-tuning, architecture design of foundation model based systems, and responsible-AI-by-design. This taxonomy provides concrete guidance for making major design decisions when designing foundation model based systems and highlights trade-offs arising from design decisions.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/13/2023

Towards Responsible AI in the Era of ChatGPT: A Reference Architecture for Designing Foundation Model-based AI Systems

The release of ChatGPT, Bard, and other large language model (LLM)-based...
research
08/11/2023

Decentralised Governance for Foundation Model based AI Systems: Exploring the Role of Blockchain in Responsible AI

Foundation models including large language models (LLMs) are increasingl...
research
03/23/2023

Design Patterns for AI-based Systems: A Multivocal Literature Review and Pattern Repository

Systems with artificial intelligence components, so-called AI-based syst...
research
03/17/2021

Characterizing Technical Debt and Antipatterns in AI-Based Systems: A Systematic Mapping Study

Background: With the rising popularity of Artificial Intelligence (AI), ...
research
07/31/2023

To Classify is to Interpret: Building Taxonomies from Heterogeneous Data through Human-AI Collaboration

Taxonomy building is a task that requires interpreting and classifying d...
research
01/14/2021

A Tensor-Based Formulation of Hetero-functional Graph Theory

Recently, hetero-functional graph theory (HFGT) has developed as a means...
research
03/08/2019

A Taxonomy for Understanding the Security Technical Debts in Blockchain Based Systems

Blockchain is a disruptive technology intended at implementing secure de...

Please sign up or login with your details

Forgot password? Click here to reset