CDM v5.0 — Predictions falsifiables

Read full text Summarise with VIRAT
Authors: Xavier Baloche
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

This preregistration documents falsifiable predictions derived from a unified computational framework (CDM v5.0) developed by Xavier Baloche. The framework generates quantitative predictions across multiple scientific domains from structural input parameters alone. Seven core predictions (P1–P7) are registered BEFORE experimental validation, covering: (1) transcriptional burst kinetics in DNA, (2) cathode voltage prediction in battery materials, (3) structural classification of transition metal oxides, (4) stretched exponential relaxation in chalcogenide glasses, (5) hematopoietic stem cell dormancy ratios, (6) cancer cell transcriptomic signatures, and (7) topological insulator classification. Each prediction specifies quantitative success/failure criteria. The CDM methodology is proprietary and not disclosed here — only its testable outputs are registered to establish temporal priority.

Fields of Study & Subjects

Machine Learning in Materials Science Advancements in Battery Materials Interdisciplinary Research and Collaboration

Publication details

Publication date
1 January 2026
Language
EN
Document type
Preprint