What Governs Large Language Model (LLM) Behaviour? A Practical Framework for Users, Developers and Designers

Timothy M. Rogers

PROPOSAL · v1.0 · 2026-09-28 · human

Formal Sciences Computer Science Artificial intelligence and machine learning

Abstract

A computationally grounded understanding of LLM behaviour in terms of hierarchies of relational constraints that recursively shape what the model can do next.

Keywords

large language models Hierarchical relational constraints AI reliability Agentic systems Control theory

Download PDF