Neurosymbolic Architectural Reasoning: Towards Formal Analysis through Neural Software Architecture Inference
Steffen Herbold, Christoph Knieke, Andreas Rausch, Christian Schindler
TL;DR
This work addresses the gap between formal software architecture analysis and everyday software development by proposing neurosymbolic architectural reasoning that uses neural inference to produce formal architectures for symbolic verification. It outlines a six-question research agenda and discusses challenges in inferring architectures from code and design documents, ensuring alignment with intended architectures, and performing formal reasoning under uncertainty. A proof-of-concept demonstrates that LLMs can learn a symbolic architectural rule from code, though soundness and generalization remain open. If successful, this approach could enable broad, automated verification of architectural constraints, improving safety, security, and maintainability across software projects.
Abstract
Formal analysis to ensure adherence of software to defined architectural constraints is not yet broadly used within software development, due to the effort involved in defining formal architecture models. Within this paper, we outline neural architecture inference to solve the problem of having a formal architecture definition for subsequent symbolic reasoning over these architectures, enabling neurosymbolic architectural reasoning. We discuss how this approach works in general and outline a research agenda based on six general research question that need to be addressed, to achieve this vision.