Structural Inhibition of Hallucination in Large Language Models via Axiomatic Constraint Protocol (ACP):From Probabilistic Generation to Deterministic Understanding
Abstract
This paper proposes an axiomatic constraint framework for LLM reasoning, introducing protocols for logical consistency, internal time, and semantic control. It discusses safety failure modes (e.g., hallucination) and outlines mechanisms for controlled reasoning and multi-agent coordination based on inhibition/withholding states rather than output sharing. Version 2: corrected references (removed non-public citation). No changes to core claims.
Fields of Study & Subjects
Semantic Web and Ontologies
Logic
Reasoning
and Knowledge
Constraint Satisfaction and Optimization
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 6 January 2026
- Language
- EN
- Document type
- Preprint