Beyond Monolithic Models: Symbolic Seams for Composable Neuro-Symbolic Architectures

TL;DR

This paper advocates for a composable neuro-symbolic AI architecture using symbolic seams, enabling transparent, adaptable, and verifiable systems by breaking down monolithic models into inspectable, interchangeable components.

Contribution

It introduces the concept of symbolic seams as explicit architectural breakpoints to facilitate modular, transparent, and evolvable neuro-symbolic AI systems.

Findings

Seams enable combining data-driven and symbolic approaches effectively.

They improve system transparency and verifiability.

Seams support extensibility and principled evolution of AI architectures.

Abstract

Current Artificial Intelligence (AI) systems are frequently built around monolithic models that entangle perception, reasoning, and decision-making, a design that often conflicts with established software architecture principles. Large Language Models (LLMs) amplify this tendency, offering scale but limited transparency and adaptability. To address this, we argue for composability as a guiding principle that treats AI as a living architecture rather than a fixed artifact. We introduce symbolic seams: explicit architectural breakpoints where a system commits to inspectable, typed boundary objects, versioned constraint bundles, and decision traces. We describe how seams enable a composable neuro-symbolic design that combines the data-driven adaptability of learned components with the verifiability of explicit symbolic constraints -- combining strengths neither paradigm achieves alone. By treating AI systems as assemblies of interchangeable parts rather than indivisible wholes, we outline a direction for intelligent systems that are extensible, transparent, and amenable to principled evolution.