Annotation Cascading: Hierarchical Model Routing, Topical Authority, and Inter-Page Context Propagation in Large-Scale Web Content Classification

Annotation Cascading (v2, tightened — Feb 2026)
This preprint proposes a unified model for web-scale content annotation under compute constraints. It introduces Annotation Cascading (site→category→page→chunk routing through specialist models), formalises topical authority as measurable annotation routing efficiency across Closeness, Strength, and Longevity, and proposes the Navigation Memory hypothesis (inter-page context propagation via Carry vs. Break Anchors). The paper provides testable predictions, proxy metrics, and experimental designs for empirical validation.

Version: v2 (tightened), February 2026
Suggested citation: Barnard, J. (2026). Annotation Cascading: Hierarchical Model Routing, Topical Authority, and Inter-Page Context Propagation in Large-Scale Web Content Classification (Preprint, v2). Zenodo. DOI: https://doi.org/10.5281/zenodo.18723669
AI-use disclosure: Drafting and structural editing assistance was provided by Claude (Anthropic). All technical content, terminology, and analysis are the author’s.

This paper is TKF-2-18723668 in The Kalicube Framework series by Jason Barnard (Kalicube). Canonical citation identifiers for the whole research programme are maintained in The Kalicube Framework Programme Register (TKF-0): https://doi.org/10.5281/zenodo.20645889. Cite this paper as TKF-2-18723668 (Zenodo concept DOI 10.5281/zenodo.18723668).