DataGov-Factory: An Empirical Evaluation of Managed vs. Custom LLM Agent Architectures for Financial Data Governance
This paper presents the first controlled empirical comparison of three LLM deployment architectures for financial data governance: (A) platform-managed agents, (B) custom multi-agent factory, and (C) direct API calls. Key finding: platform-level prompt caching reduces context costs by ~90%, making managed agents competitive while delivering 16% higher quality. v2: Fixed broken LaTeX references.
Authors
- Karl J. Mollan Neyra (ORCID: https://orcid.org/0000-0001-8361-4638)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-04-11
- DOI
- https://doi.org/10.5281/zenodo.19520901
- Primary Topic
- Financial Reporting and XBRL
- Type
- article
- Field-Weighted Citation Impact
- 0.00