Scoped, checkable answers
Research question
Does a structured step in front of retrieval make answers over a company's systems more complete, correct and efficient?
Why it is hard
A question such as "which open orders above 10 kW, by customer?" needs every member, once, in the right unit and current. Search finds passages, not complete sets, and language models count unreliably. Retrieval also weakens as the index grows.
Hypotheses
We test whether the combined path:
- H1answers questions over sets more completely and correctly;
- H2lowers cost and time per usable answer;
- H3keeps its gain as sources and data grow;
- H4does no worse on other questions (non-inferiority).
What we measure
Share of usable results within budget, correctness, completeness (item F1, exact counts), correct numbers and units, cost and time per usable answer, traceability, and scaling as sources and data grow.
Built on
- Data quality and trust
- Distributed synchronisation
- Entity resolution and typed values
- Quasi-deterministic agents and evaluation
Studies R1 to R5, R7, R8.
Further reading
- Wong, R.; Wang, J. et al. (2025): WideSearch: Benchmarking Agentic Broad Info-Seeking. Preprint, arXiv 2508.07999.
- Ravnik, J. et al. (2026): FinSheet-Bench. Preprint, arXiv 2603.07316.
- Reimers, N.; Gurevych, I. (2021): The Curse of Dense Low-Dimensional Information Retrieval for Large Index Sizes. ACL-IJCNLP 2021.
- Gao, L. et al. (2023): PAL: Program-aided Language Models. ICML 2023.




