Paper Overview
Research Area: AI/ML Authors: MasterControl AI Lab Published: 2026-09-06 arXiv: 2509.00002
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.
Key Findings
- 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 episodes.
- The authors emphasize this is a result of the specific configurations tested and does not prove that runtime agents cannot succeed under other designs.
Original Abstract
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.
*Auto-collected on 2026-09-06.*