Feasibility-First Governance-Constrained LLM Routing: Reproducibility Package

Reproducibility package for the study "Feasibility-First Routing for Regulated Multi-Provider LLM Serving: Governance, Caching, and the Limits of Online Learning." It contains the governance-gate and routing/analysis code, the measured multi-provider inference records, figures, hypothesis documentation, an auditable decision-log sample, and the manuscript PDF. The study measures a governance gate that holds simulated-policy residency/BAA violations at 0% versus 38.5% for governance-blind routing, an exact-match policy-scoped cache as the dominant cost lever, and an honest accuracy-cost trade-off for online (LinUCB) routing (-3.45 percentage points accuracy across 40 repeated benchmark re-splits while usually lowering modeled cost). All results reproduce from the saved records with numpy only; no new API calls are required. The code is licensed MIT; the package, text, figures, and data are licensed CC BY 4.0.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-29
DOI
https://doi.org/10.5281/zenodo.23772362
Primary Topic
Ethics and Social Impacts of AI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Feasibility-First Governance-Constrained LLM Routing: Reproducibility Package

Zane Zimmerman
2 links
Zenodo (CERN European Organization for Nuclear Research)
Ethics and Social Impacts of AI
article

Feasibility-First Governance-Constrained LLM Routing: Reproducibility Package

Zane Zimmerman
article en

Abstract

Reproducibility package for the study "Feasibility-First Routing for Regulated Multi-Provider LLM Serving: Governance, Caching, and the Limits of Online Learning." It contains the governance-gate and routing/analysis code, the measured multi-provider inference records, figures, hypothesis documentation, an auditable decision-log sample, and the manuscript PDF. The study measures a governance gate that holds simulated-policy residency/BAA violations at 0% versus 38.5% for governance-blind routing, an exact-match policy-scoped cache as the dominant cost lever, and an honest accuracy-cost trade-off for online (LinUCB) routing (-3.45 percentage points accuracy across 40 repeated benchmark re-splits while usually lowering modeled cost). All results reproduce from the saved records with numpy only; no new API calls are required. The code is licensed MIT; the package, text, figures, and data are licensed CC BY 4.0.

Zenodo (CERN European Organization for Nuclear Research)
Texas A&M University (US)
Openalex Percentile: Top 8%
Ethics and Social Impacts of AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon