SemPlan Benchmarks LLM Queries Over Enterprise Data

Bruno Santos Teixeira· August 17, 2026 View original

Key takeaways

  • LLM-based queries over enterprise data face challenges in correctness and governance.
  • SemPlan is a new benchmark for evaluating structured semantic planning architectures.
  • No single architecture universally outperforms others; trade-offs exist.
  • Structured semantic planning can improve correctness, but policy compliance and cost vary.

Who benefits

Enterprise SoftwareData AnalyticsFinancial ServicesHealthcareAI/Tech

Summary

SemPlan is a new bilingual benchmark evaluating structured semantic planning for LLM-based queries over enterprise data, comparing four architectural designs. Results show overall low correctness but highlight trade-offs between correctness, policy compliance, cost, and ambiguity resolution across different approaches.

Natural language interfaces for querying enterprise data face significant challenges in translating ambiguous user requests into executable, governed actions while managing costs and ensuring policy compliance. The SemPlan Benchmark was developed to evaluate different architectural approaches for this problem. It's a deterministic, synthetic bilingual benchmark with 1,800 cases, designed to test how well LLM-based systems can handle structured semantic planning. The study compared four architectures: direct SQL generation, a bounded tool-agent, structured semantic-request generation with deterministic planning, and a clarification-aware semantic-plan variant. Absolute answer correctness was generally low across all architectures, with the structured semantic-request approach showing the highest correctness, though not dramatically. However, different architectures excelled in specific areas, such as policy compliance or cost efficiency, indicating a trade-off rather than a universally superior solution.

Why it matters

Professionals building or integrating LLM-based natural language interfaces for enterprise data need to understand the trade-offs and limitations of different architectural designs to ensure accuracy, compliance, and cost-effectiveness.

How to implement this in your domain

  1. 1Utilize the SemPlan Benchmark to evaluate the performance of your LLM-based data querying systems.
  2. 2Consider architectural designs that prioritize structured semantic planning to improve correctness, even if overall accuracy remains challenging.
  3. 3Implement robust policy enforcement mechanisms, as direct SQL generation showed higher policy-correct rates.
  4. 4Analyze the trade-offs between correctness, cost, and ambiguity resolution when selecting or designing your LLM query architecture.

Original post by Bruno Santos Teixeira

"arXiv:2608.13612v1 Announce Type: new Abstract: Natural-language interfaces to enterprise data must translate underspecified requests into governed, executable behavior while controlling invalid queries, policy failures, cost, and nondeterminism. SemPlan Benchmark evaluates this…"

View on X

Originally posted by Bruno Santos Teixeira on X · view source

Want to go deeper?

Turn these trends into skills with Learnijoy's hands-on AI & tech courses.

Explore courses