Theory as a Model-Generating System II: Iterated Collapse, Evaluative Capacity, and Self-Disrupting Workflow Policies in Human–LLM Reasoning
Abstract
Preprint v1.1. This paper studies the dynamics induced when staged reduction operators are applied repeatedly in a human–LLM generator–selector workflow under budget constraints. It introduces evaluative capacity as a human-centered operational state variable and proposes self-disrupting workflow policies (SDWP) implemented above reduction operators. Empirical anchoring reports large-scale simulations demonstrating proxy-trap lock-in and consistent recovery under SDWP.
Fields of Study & Subjects
Complex Systems and Decision Making
Complex Systems and Dynamics
Artificial Intelligence Applications
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 13 February 2026
- Language
- EN
- Document type
- Preprint