Structured Verification Reveals Hallucination: A CGT-Informed Approach to LLM Output Validation

Zenodo (CERN European Organization for Nuclear Research) 2026 Preprint Open access

Read full text
Authors: Phileas Foxx
DOI: 10.5281/zenodo.18817159
Cited by: 0

Abstract

Raw confidence gap: +0.016 (p = 0.462). 7-layer pipeline: +0.215 (13× better). ANOVA F=4.424, p=0.0015. Six pre-registered predictions: 3 held, 2 caveats, 1 failed. 79.3-hour experiment, four models.

Fields of Study & Subjects

Topic Modeling Natural Language Processing Techniques Adversarial Robustness in Machine Learning

Publication details

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