MasterControl: Governed Enterprise Analytics with Policy-Executed Programs Instead of Runtime SQL Agents

Forum topic · 小凯 · 2026-09-06

Summary

This arXiv paper (2509.00002) from MasterControl AI Lab studies a governed approach to enterprise analytics in which a language model interprets the user's question, while a deterministic policy selects and runs a pre-approved analytical program that returns both results and evidence. The authors show this restriction remains expressive within a defined analytical class, using relational operations plus aggregation, comparison, windows, ranking, and similarity. Fixed semantics, policy, data, and execution rules make results fully replayable. Across 440 runs, three 8B models generated SQL and selected tools at runtime, while Qwen3-8B interpreted intent only and policy executed the approved program. None of 330 runtime-planning episodes matched the full answer-and-evidence contract across all test datasets, whereas the policy-executed analyzer matched 110 of 110. The authors note this reflects their specific configuration and does not prove runtime agents cannot succeed under other designs. The post includes a Chinese summary of the paper and links to the arXiv abstract.

Paper Overview

Research Area: AI/ML Authors: MasterControl AI Lab Published: 2026-09-06 arXiv: 2509.00002

English Summary

We study a governed approach to enterprise analytics: a language model interprets the question, while deterministic policy selects and runs a pre-approved analytical program that returns both results and evidence.

We show that this restriction can remain expressive within a defined analytical class, using relational operations plus aggregation, comparison, windows, ranking, and similarity. Fixed meaning, policy, data, and execution rules also make results replayable.

Across 440 runs, three 8B models generated SQL and selected tools at runtime, while Qwen3-8B interpreted intent only and policy executed the approved program. None of 330 runtime-planning episodes matched the full answer-and-evidence contract across all test datasets; the policy-executed analyzer matched 110 of 110. This is a result of the specific configuration and does not prove that runtime agents cannot succeed under other designs.

Key Takeaways

LLM handles intent interpretation; deterministic policy selects and executes pre-approved analytical programs.
The constrained analytical class stays expressive via relational operations, aggregation, comparison, windows, ranking, and similarity.
Fixed semantics, policy, data, and execution rules make results replayable.
Empirical comparison: 0 of 330 runtime-planning episodes matched the full answer-and-evidence contract, versus 110 of 110 for the policy-executed analyzer (across 440 total runs with three 8B models, including Qwen3-8B).
The authors explicitly scope the claim: results apply to this configuration only.

---

*Auto-collected on 2026-09-06.*