Epistemic Control Barrier Functions for LLM Agents: A Formal Framework for Verified Decision Governance

Read full text Summarise with VIRAT
Authors: Elisabeth Steurer, Gerhard Hirschmann
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

We introduce Epistemic Control Barrier Functions (Epistemic CBF), a formal framework extending Control Barrier Functions from physical safety to epistemic safety for LLM agents. The DDGK (Deterministic Decision Governance Kernel) implements this framework achieving 0% False Allow Rate across 1,200+ agent decisions. We further introduce Legal Barrier Functions for ethical/legal compliance. No prior art found for this CBF formulation applied to LLM epistemic states.

Fields of Study & Subjects

Formal Methods in Verification Safety Systems Engineering in Autonomy Ethics and Social Impacts of AI

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
25 March 2026
Language
EN
Document type
Preprint